VIU7: Sifive hat ersten RISC-V-Kern mit Vector-Erweiterung
Mit dem VIU7 bringt Sifive einen Linux-tauglichen RISC-V-Core, der dank Vector-Einheiten deutlich schneller als ARMs Cortex-A53 sein soll. (RISCV, Prozessor)
Quelle: Golem
Mit dem VIU7 bringt Sifive einen Linux-tauglichen RISC-V-Core, der dank Vector-Einheiten deutlich schneller als ARMs Cortex-A53 sein soll. (RISCV, Prozessor)
Quelle: Golem
Am Dienstag schloss Schweden Huawei und ZTE aus, am darauffolgenden Mittwoch hat Ericsson Rekordergebnisse wegen seiner Verträge in China gemeldet. (Ericsson, Huawei)
Quelle: Golem
Cloud Run now sends a SIGTERM signal to your container instance before the container instance terminates, due to an event like scale down or deleted revision. By handling this signal, you can now gracefully terminate your applications and do some cleanup tasks –as opposed to an abrupt shutdown of the container.In this blog, we will explore some use cases for this and how you can try it out.Graceful shutdownsWhen a container instance is shut down on Cloud Run, a SIGTERM signal will be sent to the container and your application will have 10 seconds to exit. If the container does not exit by then, a SIGKILL signal (which you cannot capture) will be sent to abruptly close your application. If you choose not to write a signal handler for SIGTERM, your process is terminated instantly.Using this termination signal, you can perform various “graceful shutdown” tasks in your application code:Flush monitoring data: If you use Cloud Trace or upload metrics from your application, you can develop a signal handler and call the function that flushes out the trace spans collected before your container quits and loses these in-memory trace spans that are not uploaded.Log termination of your container: By logging the termination event of the container, you can refer to your application logs to see when a specific container instance has started and exited, and get full visibility into the lifecycle of individual container instances.Close file descriptors or database connections: Some abruptly quit connections can confuse the connected servers and cause them to keep connections open for a long time than gracefully disconnecting.The graceful termination signal is primarily sent to your application when it’s scaling down container instances that are not getting traffic. Therefore, you don’t need to handle draining in-flight requests in your signal handler. However, you might sometimes receive this signal before your container will be shut down due to underlying infrastructure reasons and your container might still have in-flight connections. The graceful termination is therefore not always guaranteed.Trapping signals, the right wayMost programming languages provide libraries to trap termination signals like SIGTERM and run routines before your program terminates.If your application does not receive the termination signal on Cloud Run, the most prominent reason for this might be because your application is not running as the init process (PID 1) and its parent process is not forwarding the signal appropriately.The leading reason why this happens is that the ENTRYPOINT statement in your container image’s Dockerfile is not set directly to your application process. For example, the Dockerfile statement:ENTRYPOINT node server.jsinternally is translated to:ENTRYPOINT [“/bin/sh”, “-c”, “node server.js”]when your Dockerfile is executed to build a container image.Most notably, the GNU /bin/sh and other shells like bash do not forward signals to child processes by default. Therefore, you should write your entrypoint statements in the vector form, like the following, to prevent your app to be executed as the sub-process of a shell:ENTRYPOINT [“python”, “server.js”]Similarly, if you use an entrypoint script to kick off background processes in your containers, consider using a proper init process that can forward signals to child processes, such as tini, dumb-init or supervisord. (I have compared init process alternatives for the multi-process container use case here.)See graceful shutdowns in actionTo try trapping SIGTERM signals in Cloud Run, let’s a small Node.js server app to try this out. For this purpose, let’s take the Node.js sample application for Cloud Run. You can download it from this repository on GitHub.Add this snippet of code to index.js:process.on(‘SIGTERM’, function () { console.log(‘helloworld: received SIGTERM, exiting gracefully’); process.exit(0); });After completing this step, you can now build and push this container image, and deploy it to Cloud Run. As part of your new deployment a container instance is spun up to handle the request.After some time passes, your container will scale to zero since it is not getting any requests. (If you want to trigger a scale-to-zero event, you can also edit your Cloud Run application’s settings such as CPU or memory on the Google Cloud Console. This will deploy a new revision, and the old revision will be turned off.)The scale-to-zero will trigger a SIGTERM signal to be sent to your container before it is shut down, and you can see the graceful shutdown routine executed in the Logs tab:As you can see, Cloud Run does not require additional settings to enable graceful shutdowns. It’s turned on by default for all Cloud Run services. ConclusionIf you have cleanup tasks or have monitoring data to push out before your serverless container instances on Cloud Run shut down, give termination signals a try. Check out our documentation for more information about this feature.
Quelle: Google Cloud Platform
Belgian fintech company, Unifiedpost Group has deployedProcurement DocAI to process nearly 350 million invoices and other procure-to-pay docs per year, in 15 countries across Europe. Procurement DocAI, which was announced at Google Cloud Next OnAir, automates the capture of invoices, receipts and other procurement data at scale. It takes unstructured documents across a variety of formats and turns them into cleanly structured data, increasing operational efficiency and improving document processing accuracy. “At Unifiedpost, we believe that administrative and financial processes should be simple and smart. Google Cloud’s Procurement DocAI solution helps us achieve that goal by providing the best in class document automation processing with high accuracy and global, multi-language support.” – Hans Leybaert, CEO, Unifiedpost GroupLarge enterprise procurement and distribution networks as well as small and medium enterprises (SMEs) generate millions of invoices, receipts, and other related documents a year. These processes generate significant overhead for every procured item. With competitive pressures increasing, businesses are finding new ways to automate one of their highest volume business processes — the procurement cycle.A key part of our strategy at Google Cloud is the creation of industry-specific solutions that address vertical needs. Cross-industry solutions like Document AI are built to plug into your existing workflows and deliver business results from AI without having to hire an army of AI experts, or manage cloud infrastructure to get there.Procurement DocAI is one of our newest solutions, and it’s deployed by many customers and partners including Unifiedpost. The company currently serves 400k SMEs and 250+ corporations; their offering includes: documents (e.g., invoice PDF to XML conversion, e-invoicing), identity (e.g., KYC), and payments (e.g., IBAN accounts, online collection services, PSD2 payment services). With a large, rapidly growing, and heavily multi-language footprint, Unifiedpost sought the best industry solutions to meet their automation challenges and connect their customers such as Billtobox.com or JeFacture.com. Google Cloud’s Procurement DocAI delivered two key benefits to address Unifiedpost’s business needs:Lower TCO of procure-to-pay processing. Unifiedpost will be able to lower their TCO of procure-to-pay processing costs by up to 60% with Procurement DocAI. This solution provided them a cost effective approach for data extraction for invoices, receipts, and other valuable documents in the procurement cycle of procure to invoice, and invoice to pay.Improve procurement document processing accuracy. Procurement DocAI also helped boost data accuracy by 250% for Unifiedpost’s document extraction through specialized DocAI parsers with advanced OCR, computer vision, and Natural Language Processing. Additionally, Unifiedpost needed rapid multi-language expansion to enable regionalization support across Europe, particularly for French and Dutch to begin with, which Google delivered in less than a month.The collaboration between Google Cloud and Unifiedpost is just one of the latest examples of how we’re providing AI-powered functional solutions to solve business problems by leveraging our Deployed AI approach.Let’s connectFor more customer stories from EMEA, check out Google Cloud Next OnAir EMEA, which has tailored content to support the needs of our customers in Europe, Middle East, and Africa, kicking off from 29 September – 27 October.
Quelle: Google Cloud Platform
DJI hat mit der Pocket 2 eine kleine Kamera mit Gimbal vorgestellt, die bessere Aufnahmen machen soll als der Vorgänger. (DJI, Foto)
Quelle: Golem
Kritische Sicherheitslücken, die Golem.de dem Entwickler der Videochat-Software Big Blue Button meldete, sind erst nach Monaten geschlossen worden. Eine Recherche von Hanno Böck (Big Blue Button, Ubuntu)
Quelle: Golem
Untere Mittelklasse bis High-End: Die PCIe-Gen4-Controller von Silicon Motion werden bald in vielen NVMe-SSDs stecken. (Solid State Drive, Speichermedien)
Quelle: Golem
Experimentierfreudige können sich den Patch per Windows Update herunterladen. Er bringt ein neu gestaltetes Startmenü in Windows 10. (Windows 10, Microsoft)
Quelle: Golem
Apple hat mit iOS 14.1 sein Betriebssystem für iPhones aktualisiert und Fehlerbehebungen und Detailverbesserungen vorgenommen. (Betriebssystem, iPhone)
Quelle: Golem
One of the more critical, but sometimes forgotten, questions related to data security is how do you find the data you need to secure. To security newcomers, this may sound contradictory; surely you have that valuable data firmly in your hands?In reality, many types of sensitive, personal and even regulated data get “misplaced” at some organizations. For example, cases where payment data—credit card numbers, in particular—is found outside of the formally defined Cardholder Data Environment (CDE) have been strikingly common over many years. Sadly, they often come to light during a post-breach investigation or, somewhat better, during a PCI DSS assessment by a QSA.Similarly, recent attention (such as due to GDPR or CCPA) paid to personally identifiable information has led to cases where personal data was discovered in unexpected places. Furthermore, the accelerating pace of cloud migrations means that there are more cases of personal data being uploaded to the public cloud. It happens sometimes without the necessary controls, and, in fact, without awareness of security and privacy teams. For example, a test instance of a data analysis application may be moved from the data center to the cloud, without thinking that the test instance used production customer data. In fact, perhaps it was acceptable to use personal data for testing while the application was developed and then deployed internally, but now public cloud changed things.These and other similar cases have elevated the importance of data discovery, a key component of DLP technology. As we noted in our previous blog, sensitive data discovery is critically important for security, compliance and privacy initiatives. Thus, there is value in knowing where your sensitive data is at any time, whether it is in the cloud or not.Perhaps surprisingly, one can still see situations where sensitive data discovery is a “hard sell” with security leaders. Some leaders see the value in preventing the leaks (and theft) of valuable data across the perimeter, but not necessarily the discovery of the data inside the perimeter. However, the fact is that such thinking has become outdated in the cloud era! The perimeter has morphed in many ways hence simply sitting at the border (that is, if you can find the border to sit on) looking for departing data is no longer real (that is, if you assume that it ever was). In light of this, there are organizations that consider a broad accidental disclosure of sensitive data inside their organization to be “an internal data breach”, even though the data was never seen departing from the company. In fact, in a global organization, such internal disclosure may violate rules because it may make the data visible by employees from other countries.Why discover?Hence, the only approach that works today is protecting sensitive data by starting with knowing where it exists. This may have been conceptually true for years, but today this is also true operationally. Cloud has made this true!Still, there is a substantial debate about sensitive data in the cloud. One survey found out that “71% of organizations report that the majority of their cloud-resident data is sensitive.” However, the real challenge is that it is very likely that many organizations have sensitive data in the cloud and they are not aware of what data and where in the cloud. Gartner recently noted that data discovery plays a role in Data Access Governance (DAG). Hence, even though discovery on its own does not make the data “more secure”, it is a critical first step to take. It can make decisions about the data (approving access requests, sharing, retention, etc.) more informed and thus more secure.What to discover?The definition of sensitive data remains the subject of some debate in the security community. Some define it as data that, if revealed, will cause harm; some focus on data that others may want to steal; and some use the pure regulatory definition (hence substituting “regulated” data for “sensitive”—perhaps not a very logical change).Still, there are some types of data where there is broad agreement that such data is considered sensitive (even though the universal definition of “sensitive data” perhaps remains elusive):Regulated data such as payment data, personally identifiable information (PII) and many types of personal health information (PHI).Corporate secrets and other data that is sensitive because it is clearly valuable for business.Data that if made public will cause harm, negative PR or other damage to a company and/or its brand.It is very likely that entire industries and even specific companies can identify many other types of data considered sensitive. Note that valuing data as a business asset is an area of much research.When to discover?Our conversation here focuses on sensitive data in the cloud, hence it is useful to relate our discovery activities to cloud migration. Sensitive data discovery has value across the entire migration process.Before cloud migration—this helps plan what data can be moved to the public cloud and whether additional controls will be needed when it moves to specific cloud services. This ultimately helps organizations make an informed decision about sensitive data in the cloud.During cloud migration—this focuses on validating that the data being migrated is being moved into the properly secured areas. It also checks for mistakes with data classification (e.g. moving secret data to an open environment by mistake or moving regulated data into an environment without the prescribed controls). This may also be used to drive data transformation (masking, tokenization, de-identification) for reducing the risk.After cloud migration—this looks for mistakes in placing the data, moving data from more protected to less protected areas by mistake, and many other user cases. This activity evolves into an ongoing set of discovery activities that continue indefinitely. Security and compliance implications of this may include changing permissions, moving data to more protected areas and of course encrypting it.To migrate and operate sensitive data workloads in the cloud, you would very likely utilize a combination of all three of the above.How to discover?Based on when you’re performing data discovery, there are a few practices to consider.Before cloud migration, you would scan specific locations and systems to be migrated, likely looking for specific data types. Inspecting data before migration can also help inform how you will migrate the data. For example, will you triage certain data to stay off-cloud and other data to be cleared to move to cloud? Or will you employ a de-identification strategy to selectively mask and tokenize sensitive data as it migrates? During migration, data being migrated is filtered through a DLP engine looking for sensitive data. Here are two examples of where you can use DLP during migration:Transforming data—During migration you might want to inspect for and remove or mask sensitive data. This can be a technique to lower the risk inherent to certain data types and to compliment other security controls like encryption at rest and access control. (Example migration solution)Quarantine—Let’s say that you have data migrating to the cloud from several sources (internal, partners, customers, etc.) and you are not able to always inspect the data ahead of time. You can have this data land in a protected zone first and then use DLP to scrutinize it. After that, based on inspection results, you can either keep the data locked or to release the data from the “lockdown” into its intended and approved location. (Example triage solution). When focused on ongoing discovery activities, it makes sense to structure the scans as broad (discover some or all types of sensitive data across all systems) or deep (discover specific types of sensitive data across one particular system). This will also serve the needs of data access governance. (Who is accessing what data and why?) The pattern “broad first, deep second” is in fact an effective way to organize discovery. For example, a broad scan of many (or all) cloud locations for many data types may answer the questions “Do you have sensitive data in the cloud?”, “What data, specifically?” and “Where in the cloud, exactly?” Such broad scans should ideally be ongoing.Finally, ad hoc scans are also part of the mix. During an audit, a scan may be run over many locations looking for a specific data type. Howerer, if a particular project is being tested for security and privacy issues, its environment may be scanned for a long list of sensitive data types.A detailed discussion of how DLP helps data transformation is coming in the next post.Resources to reviewWatch Cloud Next OnAir Hybrid DLP VideoCreate a scan of your cloud storageTry out one of our demo tutorialsActions to plan for your data security programIf you’re in the cloud, the internal / external distinction goes away, so you need to be more proactive about data governance.Moving to cloud allows you to make data discovery part of your normal BAU processes and that will aid in security, compliance, and governance. Include pre- and post-migration data discovery efforts to your program or verify the ongoing discovery activities.Related ArticleNot just compliance: reimagining DLP for today’s cloud-centric worldA look back at the history of DLP before discussing how DLP is useful in today’s environment, including compliance, security, and privacy…Read Article
Quelle: Google Cloud Platform