Kategorie: Microsoft Azure

How we interact with information: The new era of search

In today’s rapidly evolving technological landscape, generative AI, and especially Large Language Models (LLMs), are ushering in a significant inflection point. These models stand at the forefront of change, reshaping how we interact with information.

The utilization of LLMs for content consumption and generation holds immense promises for businesses. They have the potential to automate content creation, enhance content quality, diversify content offerings, and even personalize content. This is an inflection point and great opportunity to discover innovative ways to accelerate your business’s potential; explore the transformative impact and shape your business strategy today.

LLMs are finding practical applications in various domains. Take, for example, Microsoft 365 Copilot—a recent innovation aiming to reinvent productivity for businesses by simplifying interactions with data. It makes data more accessible and comprehensible by summarizing email threads in Microsoft Outlook, highlighting key discussion points, suggesting action items in MicrosoftTeams, and enabling users to automate tasks and create chatbots in Microsoft Power Platform.

Data from GitHub demonstrates the tangible benefits of Github Copilot, with 88 percent of developers reporting increased productivity and 73 percent reporting less time spent searching for information or examples.

Transforming how we search

Remember the days when we typed keywords into search bars and had to click on several links to get the information we needed?

Today, search engines like Bing are changing the game. Instead of providing a lengthy list of links, they intelligently interpret your question and source from various corners of the internet. What’s more, they present the information in a clear and concise manner, complete with sources.

The shift in online search is making the process more user-friendly and helpful. We are moving from endless lists of links towards direct, easy-to-understand answers. The way we search online has undergone a true evolution.

Now, imagine the transformative impact if businesses could search, navigate, and analyze their internal data with a similar level of ease and efficiency. This new paradigm would enable employees to swiftly access corporate knowledge and harness the power of enterprise data. This architectural pattern is known as Retrieval Augmented Generation (RAG), a fusion of Azure Cognitive Search and Azure OpenAI Service—making this streamlined experience possible.

The rise of LLMs and RAG: Bridging the gap in information access

RAG is a natural language processing technique that combines the capabilities of large pre-trained language models with external retrieval or search mechanisms. It introduces external knowledge into the generation process, allowing models to pull in information beyond their initial training.

Here’s a detailed breakdown of how RAG works:

Input: The system receives an input sequence, such as a question that needs an answer.

Retrieval: Prior to generating a response, the RAG system searches for (or “retrieves”) relevant documents or passages from a predefined corpus. This corpus could encompass any collection of texts containing pertinent information related to the input.

Augmentation and generation: The retrieved documents merge with the original input to provide context. This combined data is fed into the language model, which generates a response or output.

RAG can tap into dynamic, up-to-date internal and external data sources, and can access and utilize newer information without requiring extensive training. The ability to incorporate the latest knowledge leads to better precise, informed, and contextually relevant responses that brings a key advantage.

RAG in action: A new era of business productivity

Here are some scenarios where RAG approach can enhance employee productivity:

Summarization and Q&A: Summarize massive quantitates of information for easier consumption and communication.

Data-driven decisioning: Analyze and interpret data to uncover patterns, and identify trends to gain valuable insights.

Personalization: Tailor interactions with individualized information to result in personalized recommendations.

Automation: Automate repetitive tasks to streamline and be more productive.

As AI continues to evolve, its applications across various fields are becoming increasingly pronounced.

The RAG approach for financial analysis

Consider the world of financial data analysis for a major corporation—an arena where accuracy, timely insights, and strategic decision-making are paramount. Let’s explore how RAG use cases can enhance financial analysis with a fictitious company called Contoso.

1. Summarization and Q&A

Scenario: ‘Contoso’ has just concluded its fiscal year, generating a detailed financial report that spans hundreds of pages. The board members want a summarized version of this report, highlighting key performance indicators.

Sample prompt: “Summarize the main financial outcomes, revenue streams, and significant expenses from ‘Contoso’s’ annual financial report.”

Result: The model provides a concise summary detailing ‘Contoso’s total revenue, major revenue streams, significant costs, profit margins, and other key financial metrics for the year.

2. Data-driven decisioning

Scenario: With the new fiscal year underway, ‘Contoso’ wants to analyze its revenue sources and compare them to its main competitors to better strategize for market dominance.

Sample prompt: “Analyze ‘Contoso’s revenue breakdown from the past year and compare it to its three main competitors’ revenue structures to identify any market gaps or opportunities.”

Result: The model presents a comparative analysis, revealing that while ‘Contoso’ dominates in service revenue, it lags in software licensing, an area where competitors have seen growth.

3. Personalization

Scenario: ‘Contoso’ plans to engage its investors with a personalized report, showcasing how the company’s performance directly impacts their investments.

Sample prompt: “Given the annual financial data, generate a personalized financial impact report for each investor, detailing how ‘Contoso’s’ performance has affected their investment value.”

Result: The model offers tailored reports for each investor. For instance, an investor with a significant stake in service revenue streams would see how the company’s dominance in that sector has positively impacted their returns.

4. Automation

Scenario: Every quarter, ‘Contoso’ receives multiple financial statements and reports from its various departments. Manually consolidating these for a company-wide view would be immensely time-consuming.

Sample prompt: “Automatically collate and categorize the financial data from all departmental reports of ‘Contoso’ for Q1 into overarching themes like ‘Revenue’, ‘Operational Costs’, ‘Marketing Expenses’, and ‘R&D Investments’.”

Result: The model efficiently combines the data, providing ‘Contoso’ with a consolidated view of its financial health for the quarter, highlighting strengths and areas needing attention.

LLMs: Transforming content generation for businesses

Leveraging RAG based solutions, businesses can boost employee productivity, streamline processes and make data-driven decisions. As we continue to embrace and refine these technologies, the possibilities for their application can be virtually limitless.

Where to start?

Microsoft provides a series of tools to suit your needs and use cases.

Learn more about using your data with Azure OpenAI Service.

What is Azure Machine Learning prompt flow?

Orchestrate your AI with Semantic Kernel.

Discover a sample app for the RAG pattern using Azure Cognitive Search and Azure OpenAI.

Learn more

Check out below partner solutions for a jumpstart.

Learn about generative AI with Avanade.

Discover generative AI technology services through Accenture.

Explore EY’s AI consulting Services.

PwC provides AI everywhere.

KPMG presents speed to modern data, analytics, and AI.

Integration of RAG into business operations is not just a trend, but a necessity in today’s data-driven world. By understanding and leveraging these solutions, businesses can unlock new avenues for growth and productivity.

The post How we interact with information: The new era of search appeared first on Azure Blog.
Quelle: Azure

Microsoft and Adobe partner to deliver cost savings and business benefits

Delivering quality end-to-end digital experiences can be challenging for multiple reasons including lack of resources, legacy technologies, and disorganized customer journeys. Microsoft and Adobe have purpose-built integrations to overcome these challenges with a result of simplifying deployment and reducing overall cost.

Grounded in open software standards and a scalable, secure cloud, Microsoft and Adobe deliver end-to-end technology ecosystems for a modern, secure, and connected enterprise. The integration between our applications transforms data into insights that enable intelligent, targeted, and customized marketing campaigns. With Microsoft setting the data foundation and Adobe providing a comprehensive marketing activation layer, we can take organizations from traditional batch marketing to real-time, precise, and timely event-based marketing.

Marketing needs to be tailored specifically to how, when, and where customers want to shop. To achieve this, Microsoft and Adobe are providing tools to make collaboration between employees easier and smarter, such as digital signatures and document sharing. Meanwhile, automation tools enable organizations to source, adapt, and deliver assets for more personalized customer experiences. These sophisticated integrations between Microsoft and Adobe are enabling businesses to do more with less.

Microsoft and Adobe partner to create a connected enterprise

A 2023 commissioned study conducted by Forrester Consulting—The Total Economic Impact™ Of Adobe SaaS Solutions with Microsoft Cloud—uncovered how our partnership and technology collaboration with Adobe lowers time and resource costs, improves employee productivity and customer engagement, as well as the return on investment by deploying Adobe applications for the enterprise on the Microsoft Cloud. The study specifically focuses on Adobe Experience Cloud and Adobe Document Cloud running on Microsoft Azure, Microsoft 365, and Microsoft Dynamics 365.

The study explores the challenges organizations hope to address by implementing the integrated Microsoft and Adobe solutions (see image below).

The resounding solutions to these challenges show the strength of the native integration between Microsoft and Adobe SaaS solutions, which enabled organizations to:

Enhance customer experiences by leveraging consolidated customer data, and with cloud support across solutions, gain real-time insights and analysis about these customers and marketing efforts.

Strengthen security and protection of data files across the enterprise with tightly integrated tools. With consolidated tools under an all-in-one vendor, tool deployment and IT team management is simplified.

Streamline data access and management across the organization. With tight integrations, organizations can improve collaboration, decision-making, and performance among their teams.

Integrated data means better customer journeys

As customers move between digital channels—mobile apps, social media, online chat, and so on—they generate digital records. Capturing insights from siloed data streams can be a challenge, impacting the ability to create personalized experiences in a timely manner for customers. By transitioning companies from legacy and siloed technologies to a connected, cloud-enabled tech stack, Microsoft and Adobe address the data issues that improve the quality of connections companies have to their customers.

According to the Forrester study, those implementing Adobe SaaS solutions on Microsoft Cloud say that managing customer experiences is the top goal of their organization (71 percent). This is also where organizations found key benefits of implementing integrated solutions.

Some survey participants saw a 45 percent gain in customer loyalty that they attributed to integrating Microsoft Cloud with Adobe Experience Cloud.

Data analysis was 15 percent faster, which improved the overall speed of work for marketers by 10 percent.

Customer satisfaction, the number of transactions, and customer retention rates all increased due to the data-driven capabilities the integration presents.

AI and Machine Learning are pivotal in sifting through large volumes of data, pinpointing areas of importance or anomalies for marketers. The wide range of triggering events, including email, calendar invites, webinars, advertising, data changes, and ERP events offer a holistic view of the customer. A focus on a self-service model delivers speed and ease of use. In combination, organizations gain operational efficiencies, as well as fast, effective marketing from Microsoft and Adobe integrations.

Increase data security across the enterprise

Keeping the organizations’ data and files secure is a leading concern of enterprise leaders—especially with data spread between on-prem servers and the cloud and hybrid work arrangements. The fear is that, without tighter integration with cloud security controls, their data centers are vulnerable to attacks.

Security is inherent with the cloud. Additionally, data stored in Microsoft Cloud and imported directly into natively connected Adobe solutions with native connectors reduces data vulnerabilities. Interviewees of the Forrester study reported that data and files specific to Adobe had not experienced any compromises.

Reduce time and resource demands to streamline deployment

Without the tight integration between Microsoft and Adobe, IT teams—or consultants and contractors—need to code or build custom API connectors to implement solutions. Microsoft and Adobe have developed over 60 specific integrations that are purpose-built for common use cases, reducing the workload for IT teams in customer organizations. The Forrester study found that for the composite organization the integration of Adobe tools went from a multi-month endeavor down to one month and required far fewer resources. This led to overall IT and security team productivity going up by 20 percent. One interviewed company reallocated 15 members of their 40-person IT team to other organizational projects. The Forrester study found that a composite organization based on interviewed customers saw the following financial benefits over three years:

251 percent ROI

1.3M USD Benefits PV

925 Thousand USD Net Present Value

Because Microsoft and Adobe support global content distribution, with Microsoft having cloud storage services worldwide, organizations can increase efficiencies in creating and distributing the millions of content assets needed to create personalized customer journeys.

Read the study to learn more

The Total Economic Impact™ Of Adobe SaaS Solutions with Microsoft Cloud report provides a deep analysis of the key challenges organizations look to address when deploying Adobe SaaS solutions on Microsoft Cloud. Read the full study to understand why.

As global leaders in business solutions, Microsoft and Adobe combine the power of data and expertise in marketing to deliver innovations that are reliable, secure, and optimized to meet consumers wherever they are in the customer journey.
The post Microsoft and Adobe partner to deliver cost savings and business benefits appeared first on Azure Blog.
Quelle: Azure

Manage your big data needs with HDInsight on AKS

As companies today look to do more with data, take full advantage of the cloud, and vault into the age of AI, they’re looking for services that process data at scale, reliably, and efficiently. Today, we’re excited to announce the upcoming public preview of HDInsight on Azure Kubernetes Service (AKS), our cloud-native, open-source big data service, completely rearchitected on Azure Kubernetes Service infrastructure with two new workloads and numerous improvements across the stack. The public preview will be available for use on 10/10.

HDInsight on AKS amplifying performance

HDInsight on AKS includes Apache Spark, Apache Flink, and Trino workloads on an Azure Kubernetes Service infrastructure, and features deep integration with popular Azure analytics services like Power BI, Azure Data Factory, and Azure Monitor, while leveraging Azure managed services for Prometheus and Grafana for monitoring. HDInsight on AKS is an end-to-end, open-source analytics solution that is easy to deploy and cost-effective to operate. 

HDInsight on AKS helps customers leverage open-source software for their analytics needs by: 

Providing a curated set of open-source analytics workloads like Apache Spark, Apache Flink, and Trino. These workloads are the best-in-class open-source software for data engineering, machine learning, streaming, and querying.

Delivering managed infrastructure, security, and monitoring so that teams can spend their time building innovative applications without needing to worry about the other components of their stack. Teams can be confident that HDInsight helps keep their data safe. 

Offering flexibility that teams need to extend capabilities by tapping into today’s rich, open-source ecosystem for reusable libraries, and customizing applications through script actions.

Customers who are deeply invested in open-source analytics can use HDInsight on AKS to reduce costs by setting up fully functional, end-to-end analytics systems in minutes, leveraging ready-made integrations, built-in security, and reliable infrastructure. Our investments in performance improvements and features like autoscale enable customers to run their analytics workloads at optimal cost. HDInsight on AKS comes with a very simple and consistent pricing structure per vcore per hour regardless of the size of the resource or the region, plus the cost of resources provisioned.

Developers love HDInsight for the flexibility it offers to extend the base capabilities of open-source workloads through script actions and library management. HDInsight on AKS has an intuitive portal experience for managing libraries and monitoring resources. Developers have the flexibility to use a Software Development Kit(SDK), Azure Resource Manager (ARM) templates, or the portal experience based on their preference.

Join us for a deep dive into this launch in our upcoming free webinar. 

Open, managed, and flexible

HDInsight on AKS covers the full gamut of enterprise analytics needs spanning streaming, query processing, batch, and machine learning jobs with unified visualization. 

Curated open-source workloads

HDInsight on AKS includes workloads chosen based on their usage in typical analytics scenarios, community adoption, stability, security, and ecosystem support. This ensures that customers don’t need to grapple with the complexity of choice on account of myriad offerings with overlapping capabilities and inconsistent interoperability.  

Each of the workloads on HDInsight on AKS is the best-in-class for the analytics scenarios it supports: 

Apache Flink is the open-source distributed stream processing framework that powers stateful stream processing and enables real-time analytics scenarios. 

Trino is the federated query engine that is highly performant and scalable, addressing ad-hoc querying across a variety of data sources, both structured and unstructured.  

Apache Spark is the trusted choice of millions of developers for their data engineering and machine learning needs. 

HDInsight on AKS offers these popular workloads with a common authentication model, shared meta store support, and prebuilt integrations which make it easy to deploy analytics applications.

Managed service reduces complexity

HDInsight on AKS is a managed service in the Azure Kubernetes Service infrastructure. With a managed service, customers aren’t burdened with the management of infrastructure and other software components, including operating systems, AKS infrastructure, and open-source software. This ensures that enterprises can benefit from ongoing security and functional and performance enhancements without investing precious development hours.  

Containerization enables seamless deployment, scaling, and management of key architectural components. The inherent resiliency of AKS allows pods to be automatically rescheduled on newly commissioned nodes in case of failures. This means jobs can run with minimal disruptions to Service Level Agreements (SLAs). 

Customers combining multiple workloads in their data lakehouse need to deal with a variety of user experiences, resulting in a steep learning curve. HDInsight on AKS provides a unified experience for managing their lakehouse. Provisioning, managing, and monitoring all workloads can be done in a single pane of glass. Additionally, with managed services for Prometheus and Grafana, administrators can monitor cluster health, resource utilization, and performance metrics.  

Through the autoscale capabilities included in HDInsight on AKS, resources—and thereby cost—can be optimized based on usage needs. For jobs with predictable load patterns, teams can schedule the autoscaling of resources based on a predefined timetable. Graceful decommission enables the definition of wait periods for jobs to be completed before ramping down resources, elegantly balancing costs with experience. Load-based autoscaling can ramp resources up and down based on usage patterns measured by compute and memory usage. 

HDInsight on AKS marks a shift away from traditional security mechanisms like Kerberos. It embraces OAuth 2.0 as the security framework, providing a modern and robust approach to safeguarding data and resources. In HDInsight on AKS authorization, access controls are based on managed identities. Customers can also bring their own virtual networks and associate them during cluster setup, increasing security and enabling compliance with their enterprise policies. The clusters are isolated with namespaces to protect data and resources within the tenant. HDInsight on AKS also allows management of cluster access using Azure Resource Manager (ARM) roles. 

Customers who’ve participated in the private preview love HDInsight on AKS. 

Here’s what one user had to say about his experience. 

“With HDInsight on AKS, we’ve seamlessly transitioned from the constraints of our in-house solution to a robust managed platform. This pivotal shift means our engineers are now free to channel their expertise towards core business innovation, rather than being entangled in platform management. The harmonious integration of HDInsight with other Azure products has elevated our efficiency. Enhanced security bolsters our data’s integrity and trustworthiness, while scalability ensures we can grow without hitches. In essence, HDInsight on AKS fortifies our data strategy, enabling more streamlined and effective business operations.” 
Matheus Antunes, Data Architect, XP Inc

Azure HDInsight on AKS resources

Learn more about Azure Kubernetes Service (AKS) 

The post Manage your big data needs with HDInsight on AKS appeared first on Azure Blog.
Quelle: Azure

Cloud Cultures, Part 3: The pursuit of excellence in the United Kingdom

The swift progression of technological innovation is truly captivating. However, for me, what holds an even greater fascination is the intricate interplay of people, narratives, and life encounters that shape how technology is used every day. The outcomes of cloud adoption are shaped dramatically by the people and their culture. These stories show firsthand how technology and tradition combine to form cloud cultures.

In our first two episodes, Poland and Sweden, we explored how the people in these countries have taken important leaps in delivering new innovative technologies, and the underlying culture that helped them get there in uniquely different ways. In Poland, there is a sense of fearlessness when it comes time to act. They are a dynamic country embracing change, reinventing themselves, and creating new innovative opportunities. For Sweden, success knows no borders. Despite being one of the largest countries in Europe by landmass, it’s one of the smallest in population, which forces Sweden’s ambitious entrepreneurs to adopt a global-first mindset from day one. Now it’s our turn to explore cloud culture in the United Kingdom. 

United Kingdom: Pursuit of excellence 

Our two data center regions in the United Kingdom have been live for nearly a decade and built to help support the exploding growth of the cloud in the United Kingdom and Europe. With the evolution of new cloud services, such as generative AI, we are on the precipice of another leap in cloud-powered innovation. My time in the United Kingdom helped me see how the innovations coming from this country and their pursuit of excellence are impacting this growth. 

In my visit to the United Kingdom, I saw how towering aspirations fuel an unwavering commitment to producing premium results and have set a global standard for greatness. Excellence is an ideal we all strive for. It drives our decision-making, fuels innovation, and propels us to exceed not only our own expectations but also the demands of an ever-evolving industry. 

The cloud culture here has shown me that excellence is more than just an outcome—it’s a mindset that permeates every aspect of our work. And while achieving it is always desirable, it’s the pursuit that matters most. 

Our conversations with customers and partners helped me see how the powerful winds of innovation that have converged with local customs, values, and ways of living, helped create something unique. 

How are United Kingdom customers using the cloud? 

These conversations helped uncover how high standards are more than just a commitment—it’s a way of life. Below are just a few of the United Kingdom customers who are transforming their businesses to adapt to the growing needs of their customers in the United Kingdom, and beyond: 

Rolls Royce Engines has delivered excellence through its engines for over 100 years. As a broad-based power and propulsion provider, they operate in some of the most complex, critical systems at the heart of global society. With each engine producing about half a gigabyte of data per flight, doing so for the next 100 years will require a transition to new ways of working. 

London Stock Exchange Group is one of the world’s leading providers of financial markets infrastructure delivering financial data, analytics, news, and index products to more than 40 thousand customers in 190 countries. Clients rely on their expertise and innovative technologies to navigate the unpredictable currents of the financial markets. And in an industry where even the slightest edge can lead to substantial margins, they’ve found theirs using the cloud to deliver insights and cutting-edge solutions at speed. 

VCreate is an innovative business operating in the sphere of healthcare. It develops secure video technology that connects patients, families, and clinical teams for improved diagnostic management and enhanced family-focused care. 

KX is a global provider of vector database technology for time-series, real-time, and embedded data that provides context and insights at the speed of thought. Its software powers generative AI applications in banking, life sciences, semiconductors, telecommunications, and manufacturing. Enabling the processing and analysis of time series and historical data at speed and scale, KX gives developers, data scientists, and engineers the tools to build high-performance data-driven applications that uncover deeper insights and drive transformative business innovation.

Data is the key 

Talking to these customers, I started to pick up on a common thread. Data is the key to unlocking this excellence. These companies process vast amounts of data in order to provide quality products or services to their customers. The cloud opens opportunities to analyze this data faster for more effective decision making, by drawing deeper insights from analytics.  

“There is a lot of focus on how to improve efficiency. You should focus more on doing the right things. It’s not about doing more for less; it’s doing the right things in the first place. It’s effectiveness, not efficiency.”—Ashok Reddy, Chief Executive Officer of KX. 

There is an important distinction between efficiency and effectiveness. Operating efficiently is undeniably important, but it doesn’t guarantee exceptional results. However, aligning our actions with meaningful outcomes can definitely be a differentiating factor. 

Learn more  

Technology is a powerful tool, not on its own but because of the people and cultures that shape it. As we move into this next era of digital transformation, with AI at the forefront, our mission has never been more important—to empower every person on the planet to achieve more. Everyone has a role to play in creating a better world and at Microsoft we simply want to provide the tools and resources to do so. 

Watch the Cloud Cultures: United Kingdom episode today.  
The post Cloud Cultures, Part 3: The pursuit of excellence in the United Kingdom appeared first on Azure Blog.
Quelle: Azure

Unlocking the potential of in-network computing for telecommunication workloads

Azure Operator Nexus is the next-generation hybrid cloud platform created for communications service providers (CSP). Azure Operator Nexus deploys Network Functions (NFs) across various network settings, such as the cloud and the edge. These NFs can carry out a wide array of tasks, ranging from classic ones like layer-4 load balancers, firewalls, Network Address Translations (NATs), and 5G user-plane functions (UPF), to more advanced functions like deep packet inspection and radio access networking and analytics. Given the large volume of traffic and concurrent flows that NFs manage, their performance and scalability are vital to maintaining smooth network operations.

Until recently, network operators were presented with two distinct options when it comes to implementing these critical NFs. One, utilize standalone hardware middlebox appliances, and two use network function virtualization (NFV) to implement them on a cluster of commodity CPU servers.

The decision between these options hinges on a myriad of factors—including each option’s performance, memory capacity, cost, and energy efficiency—which must all be weighed against their specific workloads and operating conditions such as traffic rate, and the number of concurrent flows that NF instances must be able to handle.

Our analysis shows that the CPU server-based approach typically outshines proprietary middleboxes in terms of cost efficiency, scalability, and flexibility. This is an effective strategy to use when traffic volume is relatively light, as it can comfortably handle loads that are less than hundreds of Gbps. However, as traffic volume swells, the strategy begins to falter, and more CPU cores are required to be dedicated solely to network functions.

In-network computing: A new paradigm

At Microsoft, we have been working on an innovative approach, which has piqued the interest of both industry personnel and the academic world—namely, deploying NFs on programmable switches and network interface cards (NIC). This shift has been made possible by significant advancements in high-performance programmable network devices, as well as the evolution of data plane programming languages such as Programming Protocol-Independent (P4) and Network Programming Language (NPL). For example, programmable switching Application-Specific Integrated Circuits (ASIC) offer a degree of data plane programmability while still ensuring robust packet processing rates—up to tens of Tbps, or a few billion packets per second. Similarly, programmable Network Interface Cards (NIC), or “smart NICs,” equipped with Network Processing Units (NPU) or Field Programmable Gate Arrays (FPGA), present a similar opportunity. Essentially, these advancements turn the data planes of these devices into programmable platforms.

This technological progress has ushered in a new computing paradigm called in-network computing. This allows us to run a range of functionalities that were previously the work of CPU servers or proprietary hardware devices, directly on network data plane devices. This includes not only NFs but also components from other distributed systems. With in-network computing, network engineers can implement various NFs on programmable switches or NICs, enabling the handling of large volumes of traffic (e.g., > 10 Tbps) in a cost-efficient manner (e.g., one programmable switch versus tens of servers), without needing to dedicate CPU cores specifically to network functions.

Current limitations on in-network computing

Despite the attractive potential of in-network computing, its full realization in practical deployments in the cloud and at the edge remains elusive. The key challenge here has been effectively handling the demanding workloads from stateful applications on a programmable data plane device. The current approach, while adequate for running a single program with fixed, small-sized workloads, significantly restricts the broader potential of in-network computing.

A considerable gap exists between the evolving needs of network operators and application developers and the current, somewhat limited, view of in-network computing, primarily due to a lack of resource elasticity. As the number of potential concurrent in-network applications grows and the volume of traffic that requires processing swells, the model is strained. At present, a single program can operate on a single device under stringent resource constraints, like tens of MB of SRAM on a programmable switch. Expanding these constraints typically necessitates significant hardware modifications, meaning when an application’s workload demands surpass the constrained resource capacity of a single device, the application fails to operate. In turn, this limitation hampers the wider adoption and optimization of in-network computing.

Bringing resource elasticity to in-network computing

In response to the fundamental challenge of resource constraints with in-network computing, we’ve embarked on a journey to enable resource elasticity. Our primary focus lies on in-switch applications—those running on programmable switches—which currently grapple with the strictest resource and capability limitations among today’s programmable data plane devices. Instead of proposing hardware-intensive solutions like enhancing switch ASICs or creating hyper-optimized applications, we’re exploring a more pragmatic alternative: an on-rack resource augmentation architecture.

In this model, we envision a deployment that integrates a programmable switch with other data-plane devices, such as smart NICs and software switches running on CPU servers, all connected on the same rack. The external devices offer an affordable and incremental path to scale the effective capacity of a programmable network in order to meet future workload demands. This approach offers an intriguing and feasible solution to the current limitations of in-network computing.

Figure 1: Example scenario scaling up to handle load across servers. The control plane installs programmable switch rules, which map cell sites to Far Edge servers.

In 2020, we presented a novel system architecture, called the Table Extension Architecture (TEA), at the ACM SIGCOMM conference.1 TEA innovatively provides elastic memory through a high-performance virtual memory abstraction. This allows top-of-rack (ToR) programmable switches to handle NFs with a large state in tables, such as one million per-flow table entries. These can demand several hundreds of megabytes of memory space, an amount typically unavailable on switches. The ingenious innovation behind TEA lies in its ability to allow switches to access unused DRAM on CPU servers within the same rack in a cost-efficient and scalable way. This is achieved through the clever use of Remote Direct Memory Access (RDMA) technology, offering only high-level Application Programming Interfaces (APIs) to application developers while concealing complexities.

Our evaluations with various NFs demonstrate that TEA can deliver low and predictable latency together with scalable throughput for table lookups, all without ever involving the servers’ CPUs. This innovative architecture has drawn considerable attention from members of both academia and industry and has found its application in various use cases that include network telemetry and 5G user-plane functions.

In April, we introduced ExoPlane at the USENIX Symposium on Networked Systems Design and Implementation (NSDI).2 ExoPlane is an operating system specifically designed for on-rack switch resource augmentation to support multiple concurrent applications.

The design of ExoPlane incorporates a practical runtime operating model and state abstraction to tackle the challenge of effectively managing application states across multiple devices with minimal performance and resource overheads. The operating system consists of two main components: the planner, and the runtime environment. The planner accepts multiple programs, written for a switch with minimal or no modifications, and optimally allocates resources to each application based on inputs from network operators and developers. The ExoPlane runtime environment then executes workloads across the switch and external devices, efficiently managing state, balancing loads across devices, and handling device failures. Our evaluation highlights that ExoPlane provides low latency, scalable throughput, and fast failover while maintaining a minimal resource footprint and requiring few or no modifications to applications.

Looking ahead: The future of in-network computing

As we continue to explore the frontiers of in-network computing, we see a future rife with possibilities, exciting research directions, and new deployments in production environments. Our present efforts with TEA and ExoPlane have shown us what’s possible with on-rack resource augmentation and elastic in-network computing. We believe that they can be a practical basis for enabling in-network computing for future applications, telecommunication workloads, and emerging data plane hardware. As always, the ever-evolving landscape of networked systems will continue to present new challenges and opportunities. At Microsoft we are aggressively investigating, inventing, and lighting up such technology advancements through infrastructure enhancements. In-network computing frees up CPU cores resulting in reduced cost, increased scale, and enhanced functionality that telecom operators can benefit from, through our innovative products such as Azure Operator Nexus.

References

TEA: Enabling State-Intensive Network Functions on Programmable Switches, ACM SIGCOMM 2020

ExoPlane: An Operating System for On-Rack Switch Resource Augmentation, USENIX NSDI 2023

The post Unlocking the potential of in-network computing for telecommunication workloads appeared first on Azure Blog.
Quelle: Azure

Accelerating the pace of innovation with Azure Space and our partners

Azure Space innovating into the future

Today, I’m excited to share some news spanning the full spectrum of space industry use cases, including:

Real-world examples of how Azure Orbital Ground Station is enabling both space agencies and start-ups with new ways to operate satellites in orbit.

A new addition to the Azure Space family, the Planetary Computer (and the petabytes of data within its catalog). Together, with new partnerships with Esri and Synthetaic, we are on a journey to empower our customers through rapid new insights from earth observation data. 

When Microsoft announced Azure Space in 2020, we saw a chance to innovate new solutions that meet modern demands and create opportunities for the future. So, we applied our proven partner-first approach to rapidly reimagine traditional space solutions, introduce new software-based tools, and minimize cost barriers holding back the space ecosystem.

This partner-first approach has allowed us to rapidly go from vision to solutions. In October 2020, we shared early outcomes that collectively brought Azure together with a global effort of ground stations, provided resilient connectivity to the hyperscale cloud, and launch a transformational effort to virtualize satellite communications.

Azure customers

Since then, both established satellite operators and exciting new start-ups have redefined the on-orbit possibilities. Now, start-ups, government agencies, and enterprises are experimenting with new form factor satellites, relying upon Earth observation data to derive actionable insights, and identifying ways to maintain constant connectivity in unpredictable global environments.

Together with our partners, we are rapidly innovating to provide every space operator with the solutions to solve persistent challenges in new ways and capture new opportunities in the rapidly expanding space sector.

Azure Orbital Ground Station supports our customers on and off the planet

Cloud computing is the foundation that underpins one of the biggest revolutions in the space industry, ground stations as a service. Those ground stations, in turn, have drastically lowered one of the most expensive barriers to entering space. One year after Azure Orbital Ground Station became generally available, customers including NASA and Muon Space are using it to support their operations.

Use KSAT and Azure Orbital Ground Station to improve delivery of NASA’s earth science data products

Teams from NASA (Langley Research Center and Goddard Space Flight Center), the global space company KSAT, and Microsoft have completed a technology demonstration. The demo focused on data acquisition, processing, and distribution of near real-time Earth Science data products in the cloud.

The teams successfully validated space connectivity across KSAT and Microsoft Azure Orbital Ground Station sites with four public satellites owned by both NASA and the National Oceanic and Atmospheric Administration (NOAA): Terra, Aqua, Suomi National Polar-orbiting Partnership, and JPSS-1/NOAA-20. 

The demonstration was a showcase for Azure, which provided real-time cross-region data delivery from Microsoft and KSAT sites to NASA’s virtual network in Azure. With satellite data in the cloud, NASA Azure compute and storage services to take data from raw form to higher processing levels (see sample final product in Image 1) improving latency from an average of 3-6 hours to under 25 minutes for some data products.

Integration of capabilities across Microsoft and partner KSAT allowed NASA to expand its coverage and connectivity, benefiting from ground station access through a single application programming interface (API), direct backhaul into Azure, cross-region delivery, and a unified data format experience.

Muon Space achieves liftoff and successful operations with MuSat-1 and Azure Orbital Ground Station

Muon Space selected Microsoft to support its first-ever launch in June 2023, leveraging Azure Orbital Ground Station as the sole ground station provider for their MuSat-1 mission. After MuSat-1 was deployed from the SpaceX Transporter 10, Muon Space achieved contact via Azure Orbital Ground Station within six minutes.1 From the launch and early operation (LEOP) stage to continuous on-orbit operations, Microsoft ground stations around the world are used to successfully communicate with MuSat-1. Azure Orbital Ground Station is a completely cloud-based solution. Therefore, hardware deployment was not needed, enabling Muon Space to take advantage of an innovative virtual radio frequency (RF) solution with a custom modem.

“Since we’re building constellations of multiband remote sensing spacecraft with unique revisit, resolution, and data latency capabilities, our ground station partner was a critical choice. Muon selected Azure Orbital Ground Station as we launch our constellation due to its current capabilities and product roadmap. Collaborating with Microsoft to handle ground allows us to focus on the core mission of gathering climate intelligence and serving it to our customers.”
Jim Martz, Vice President, Engineering at Muon.

Expanding capabilities to derive rapid insights from massive amounts of space data

Today, we are applying that same approach to revolutionize geospatial data, as we welcome Planetary Computer to the Azure Space family. Planetary Computer is a robust geospatial data platform. It combines a multi-petabyte catalog of global multi-source data with intuitive APIs, a flexible scientific environment that allows users to answer global questions about that data, and applications that put those answers in the hands of many stakeholders. It is used by researchers, scientists, students, and organizations across the globe with millions of API calls every month.

With Planetary Computer now part of the Azure Space family, we are beginning to work with partners in new ways. Today, we are building upon the existing catalog to drive a path to new capabilities and partnerships that will empower users with one of the largest Earth observation data sets at their fingertips—petabytes worth of possibilities for understanding our planet. In keeping with our partner-first approach, Esri and Synthetaic will provide essential capabilities to our platform and ecosystem. By combining the power of data analysis in Synthetaic’s RAIC, with the data visualization of Esri’s ArcGIS, customers will be able to glean insights from space data at previously unattainable speed and scale.

Planetary Computer is grounded in our commitment to not only better understand our world but to leverage the insights we gain to achieve our commitments to being a carbon-negative, water-positive, and zero-waste company by 2030. A quick look at Chesapeake Conservancy, a nonprofit organization based in Annapolis, Maryland, illustrates the possibilities that Planetary Computer can unlock for any organization to achieve its sustainability goals.

Chesapeake Conservancy is leading a regional effort to protect 30 percent of the Chesapeake Bay watershed by 2030 using a precision conservation approach that optimizes resources and protects land with the greatest value for water quality, outdoor recreation, wildlife habitat, and local economies.

“Collaborating with Microsoft Azure, we developed an AI system that maps ground-mounted solar arrays using up-to-date satellite data enabling us to regularly track one of the most rapid drivers of land use change in the watershed,” says Joel Dunn, CEO, Chesapeake Conservancy. “Going forward, we must contextualize these insights within a complete, up-to-date picture of land use. We’re excited to work with Microsoft and their partners to produce our 1-meter land use data more frequently and accurately keeping an active pulse on the entire Chesapeake Bay watershed.”

Using tools from our partners to harness the power of Microsoft Planetary Computer, customers such as Chesapeake Conservancy can use artificial intelligence and groundbreaking data to accelerate progress in conserving landscapes vital to the Chesapeake Bay’s health and its cultural heritage while equitably connecting people to the Chesapeake, as seen in the video here.

The space industry has made incredible progress since the dawn of the space age in the 1950s. That progress, however, has been slow, hard, and expensive. It has required expensive and sustained investments by governments, followed by the creation of bespoke, rigid, and complex-to-integrate satellite constellations and ground infrastructure.

What’s next for Azure Space

Building on the Microsoft Planetary Computer to develop a new Azure Space Data solution will create an end-to-end space fabric, providing ubiquitous connectivity, resiliency, and global insights at scale, in real-time. We are excited to build this platform and open it to the many companies, large and small, who are shaping the future of space and look forward to collaborating with those writing the next chapter in humanity’s journey beyond Earth.

We invite partners and enterprises interested in learning more about Azure Space to do the following:

Learn more about Azure Space

Sign up for news and updates on how Space data can advance your organization and missions, or complete this form to get in touch with the Azure Space team.

Save the date—join us for the Planetary Computer webinar in December.

References

Muon Space launches first satellite.

The post Accelerating the pace of innovation with Azure Space and our partners appeared first on Azure Blog.
Quelle: Azure

Microsoft and Accenture partner to tackle methane emissions with AI technology

This post was co-authored by Dan Russ, Associate Director, and Sacha Abinader, Managing Director from Accenture.

The year 2022 was a notable one in the history of our climate—it stood as the fifth warmest year ever recorded1. An increase in extreme weather conditions, from devastating droughts and wildfires to relentless floods and heat waves, made their presence felt more than ever before—and 2023 seems poised to shatter still more records. These unnerving circumstances demonstrate the ever-growing impact of climate change that we’ve come to experience as the planet continues to warm.

Microsoft’s sustainability journey

At Microsoft, our approach to mitigating the climate crisis is rooted in both addressing the sustainability of our own operations and in empowering our customers and partners in their journey to net-zero emissions. In 2020, Microsoft set out with a robust commitment: to be a carbon-negative, water positive, and zero-waste company, while protecting ecosystems, all by the year 2030. Three years later, Microsoft remains steadfast in its resolve. As part of these efforts, Microsoft has launched Microsoft Cloud for Sustainability, a comprehensive suite of enterprise-grade sustainability management tools aimed at supporting businesses in their transition to net-zero.

Moreover, our contribution to several global sustainability initiatives has the goal of benefiting every individual and organization on this planet. Microsoft has accelerated the availability of innovative climate technologies through our Climate Innovation Fund and is working hard to strengthen our climate policy agenda. Microsoft’s focus on sustainability-related efforts forms the backdrop for the topic tackled in this blog post: our partnership with Accenture on the application of AI technologies toward solving the challenging problem of methane emissions detection, quantification, and remediation in the energy industry.

“We are excited to partner with Accenture to deliver methane emissions management capabilities. This combines Accenture’s deep domain knowledge together with Microsoft’s cloud platform and expertise in building AI solutions for industry problems. The result is a solution that solves real business problems and that also makes a positive climate impact.”—Matt Kerner, CVP Microsoft Cloud for Industry, Microsoft.

Why is methane important?

Methane is approximately 85 times more potent than carbon dioxide (CO2) at trapping heat in the atmosphere over a 20-year period. It is the second most abundant anthropogenic greenhouse gas after CO2, accounting for about 20 percent of global emissions.

The global oil and gas industry is one of the primary sources of methane emissions. These emissions occur across the entire oil and gas value chain, from production and processing to transmission, storage, and distribution. The International Energy Agency (IEA) estimates that it is technically possible to avoid around 75 percent of today’s methane emissions from global oil and gas operations. These statistics drive home the importance of addressing this critical issue.

Microsoft’s investment in Project Astra

Microsoft has signed on to the Project Astra initiative—together with leading energy companies, public sector organizations, and academic institutions—in a coordinated effort to demonstrate a novel approach to detecting and measuring methane emissions from oil and gas production sites.

Project Astra entails an innovative sensor network that harnesses advances in methane-sensing technologies, data sharing, and data analytics to provide near-continuous emissions monitoring of methane across oil and gas facilities. Once operational, this kind of smart digital network would allow producers and regulators to pinpoint methane releases for timely remediation.

Accenture and Microsoft—The future of methane management

Attaining the goal of net-zero methane emissions is becoming increasingly possible. The technologies needed to mitigate emissions are maturing rapidly, and digital platforms are being developed to integrate complex components. As referenced in Accenture’s recent methane thought leadership piece, “More than hot air with methane emissions”. What is needed now is a shift—from a reactive paradigm to a preventative one—where the critical issue of leak detection and remediation is transformed into leak prevention by leveraging advanced technologies.

Accenture’s specific capabilities and toolkit

To date, the energy industry’s approach to methane management has been fragmented and comprised of a host of costly monitoring tools and equipment that have been siloed across various operational entities. These siloed solutions have made it difficult for energy companies to accurately analyze emissions data, at scale, and remediate those problems quickly.

What has been lacking is a single, affordable platform that can integrate these components into an effective methane emissions mitigation tool. These components include enhanced detection and measurement capabilities, machine learning for better decision-making, and modified operating procedures and equipment that make “net-zero methane” happen faster. These platforms are being developed now and can accommodate a wide variety of technology solutions that will form the digital core necessary to achieve a competitive advantage.

Accenture has created a Methane Emissions Monitoring Platform (MEMP) that facilitates the integration of multiple data streams and embeds key methane insights into business operations to drive action (see Figure 1 below).

Figure 1: Accenture’s Methane Emissions Monitoring Platform (MEMP).

The cloud-based platform, which runs on Microsoft Azure, enables energy companies to both measure baseline methane emissions in near real-time and detect leaks using satellites, fixed wing aircraft, and ground level sensing technologies. It is designed to integrate multiple data sources to optimize venting, flaring, and fugitive emissions. Figure 2 below illustrates the aspirational end-to-end process incorporating Microsoft technologies. MEMP also facilitates connectivity with back-end systems responsible for work order creation and management, including the scheduling and dispatching of field crews to remediate specific emission events.

Figure 2: The Methane Emissions Monitoring Platform Workflow (aspirational).

Microsoft’s AI tools powering Accenture’s Methane Emissions Monitoring Platform

Microsoft has provided a number of Azure-based AI tools for tackling methane emissions, including tools that support sensor placement optimization, digital twin for methane Internet of Things (IoT) sensors, anomaly (leak) detection, and emission source attribution and quantification. These tools, when integrated with Accenture’s MEMP, allow users to monitor alerts in near real-time through a user-friendly interface, as shown in Figure 3.

Figure 3: MEMP Landing Page visualizing wells, IoT sensors, and Work Orders.

“Microsoft has developed differentiated AI capabilities for methane leak detection and remediation, and is excited to partner with Accenture in integrating these features onto their Methane Emissions Monitoring Platform, to deliver value to energy companies by empowering them in their path to net-zero emissions”—Merav Davidson, VP, Industry AI, Microsoft.

Methane IoT sensor placement optimization

Placing sensors in strategic locations to ensure maximum potential coverage of the field and timely detection of methane leaks is the first step towards building a reliable end-to-end IoT-based detection and quantification solution. Microsoft’s solution for sensor placement utilizes geospatial, meteorological, and historical leak rate data and an atmospheric dispersion model to model methane plumes from sources within the area of interest and obtain a consolidated view of emissions. It then selects the best locations for sensors using either a mathematical programming optimization method, a greedy approximation method, or an empirical downwind method that considers the dominant wind direction, subject to cost constraints.

In addition, Microsoft provides a validation module to evaluate the performance of any candidate sensor placement strategy. Operators can evaluate the marginal gains offered by utilizing additional sensors in the network, through sensitivity analysis as shown in Figure 4 below.

Figure 4: Left: Increase in leak coverage with a number of sensors. By increasing the number of sensors that are available for deployment, the leak detection ratio (i.e., the fraction of detected leaks by deployed sensors) increases. Right: Source coverage for 15 sensors. The arrows map each sensor (red circles) to the sources (black triangles) that it detects.

End-to-end data pipeline for methane IoT sensors

To achieve continuous monitoring of methane emissions from oil and gas assets, Microsoft has implemented an end-to-end solution pipeline where streaming data from IoT Hub is ingested into a Bronze Delta Lake table leveraging Structured Streaming on Spark. Sensor data cleaning, aggregation, and transformation to algorithm data model are done and the resultant data is stored in a Silver Delta Lake table in a format that is optimized for downstream AI tasks.

Methane leak detection is performed using uni- and multi-variate anomaly detection models for improved reliability. Once a leak has been detected, its severity is also computed, and the emission source attribution and quantification algorithm then identifies the likely source of the leak and quantifies the leak rate.

This event information is sent to the Accenture Work Order Prioritization module to trigger appropriate alerts based on the severity of the leak to enable timely remediation of fugitive or venting emissions. The quantified leaks can also be recorded and reported using tools such as the Microsoft Sustainability Manager app. The individual components of this end-to-end pipeline are described in the sections below and illustrated in Figure 5.

Figure 5: End-to-end IoT data pipeline that runs on Microsoft Azure demonstrating methane leak detection, quantification, and remediation capabilities.

Digital twin for methane IoT sensors

Data streaming from IoT sensors deployed in the field needs to be orchestrated and reliably passed to the processing and AI execution pipeline. Microsoft’s solution creates a digital twin for every sensor. The digital twin comprises a sensor simulation module that is leveraged in different stages of the methane solution pipeline. The simulator is used to test the end-to-end pipeline before field deployment, reconstruct and analyze anomalous events through what-if scenarios and enable the source attribution and leak quantification module through a simulation-based, inverse modeling approach.

Anomaly (leak) detection

A methane leak at a source could manifest as an unusual rise in the methane concentration detected at nearby sensor locations that require timely mitigation. The first step towards identifying such an event is to trigger an alert through the anomaly detection system. A severity score is computed for each anomaly to help prioritize alerts. Microsoft provides the following two methods for time series anomaly detection, leveraging Microsoft’s open-source SynapseML library, which is built on the Apache Spark distributed computing framework and simplifies the creation of massively scalable machine learning pipelines:

Univariate anomaly detection: Based on a single variable, for example, methane concentration.

Multivariate anomaly detection: Used in scenarios where multiple variables, including methane concentration, wind speed, wind direction, temperature, relative humidity, and atmospheric pressure, are used to detect an anomaly.

Post-processing steps are implemented to reliably flag true anomalous events so that remedial actions can be taken in a timely manner while reducing false positives to avoid unnecessary and expensive field trips for personnel. Figure 6 below illustrates this feature in Accenture’s MEMP: the ‘hover box” over Sensor 6 documents a total of seven alerts resulting in just two work orders being created.

Figure 6: MEMP dashboard visualizing alerts and resulting work orders for Sensor 6.

Emission source attribution and quantification

Once deployed in the field, methane IoT sensors can only measure compound signals in the proximity of their location. For an area of interest that is densely populated with potential emission sources, the challenge is to identify the source(s) of the emission event. Microsoft provides two approaches for identifying the source of a leak:

Area of influence attribution model: Given the sensor measurements and location, an “area of influence” is computed for a sensor location at which a leak is detected, based on the real-time wind direction and asset geo-location. Then, the asset(s) that lie within the computed “area of influence” are identified as potential emissions sources for that flagged leak.

Bayesian attribution model: With this approach, source attribution is achieved through inversion of the methane dispersion model. The Bayesian approach comprises two main components—a source leak quantification model and a probabilistic ranking model—and can account for uncertainties in the data stemming from measurement noise, statistical and systematic errors, and provides the most likely sources for a detected leak, the associated confidence level and leak rate magnitude.

Considering the high number of sources, low number of sensors, and the variability of the weather, this poses a complex but highly valuable inverse modeling problem to solve. Figure 7 provides insight regarding leaks and work orders for a particular well (Well 24). Specifically, diagrams provide well-centric and sensor-centric assessments that attribute a leak to this well.

Figure 7: Leak Source Attribution for Well 24.

Further, Accenture’s Work Order Prioritization module using Microsoft Dynamics 365 Field Service application (Figure 8) enables Energy operators to initiate remediation measures under the Leak Detection and Remediation (LDAR) paradigm.

Figure 8: Dynamics 365 Work Order with emission source attribution and CH4 concentration trend data embedded.

Looking ahead

In partnership with Microsoft, Accenture is looking to continue refining MEMP, which is built on the advanced AI and statistical models presented in this blog. Future capabilities of MEMP look to move from “detection and remediation” to “prediction and prevention” of emission events, including enhanced event quantification and source attribution.

Microsoft and Accenture will continue to invest in advanced capabilities with an eye toward both:

Integrating industry standards platforms such as Azure Data Manager for Energy (ADME) and Open Footprint Forum to enable both publishing and consumption of emissions data.

Leveraging Generative AI to simplify the user experience.

Learn more

Case study

Duke Energy is working with Accenture and Microsoft on the development of a new technology platform designed to measure actual baseline methane emissions from natural gas distribution systems.

Accenture Methane Emissions Monitoring Platform

More information regarding Accenture’s MEMP can be found in “More than hot air with methane emissions”. Additional information regarding Accenture can be found on the Accenture homepage and on their energy page.

Microsoft Azure Data Manager for Energy

Azure Data Manager for Energy is an enterprise-grade, fully managed, OSDU Data Platform for the energy industry that is efficient, standardized, easy to deploy, and scalable for data management—ingesting, aggregating, storing, searching, and retrieving data. The platform will provide the scale, security, privacy, and compliance expected by our enterprise customers. The platform offers out-of-the-box compatibility with major service company applications, which allows geoscientists to use domain-specific applications on data contained in Azure Data Manager for Energy with ease.

Related publications and conference presentations

Source Attribution and Emissions Quantification for Methane Leak Detection: A Non-Linear Bayesian Regression Approach. Mirco Milletari, Sara Malvar, Yagna Oruganti, Leonardo Nunes, Yazeed Alaudah, Anirudh Badam. The 8th International Online & Onsite Conference on Machine Learning, Optimization, and Data Science.

Surrogate Modeling for Methane Dispersion Simulations Using Fourier Neural Operator. Qie Zhang, Mirco Milletari, Yagna Oruganti, Philipp Witte. Presented at the NeurIPS 2022 Workshop on Tackling Climate Change with Machine Learning.

1https://climate.nasa.gov/news/3246/nasa-says-2022-fifth-warmest-year-on-record-warming-trend-continues/
The post Microsoft and Accenture partner to tackle methane emissions with AI technology appeared first on Azure Blog.
Quelle: Azure