Kategorie: technology

3 ways to avoid failure in application deployment

In my previous post, I discussed six causes of system failures and their resulting negative impacts for enterprises. How can one one avoid failure?
1. If you have not yet built your DevOps practice, now is the time.
DevOps increases the feedback loop exchange from ideas to customer feedback. Automating the application release and deployment processes and using DevOps tools can drastically reduce the amount of manual intervention in the release pipeline, which leads to big gains in delivery such as faster resolution of problems, greatly reducing the complexity of problems to fix. For example, your team can use IBM Bluemix DevOps Services to develop, track, plan and deploy services in one place. This enables organizations to develop rapidly in an open, integrated environment that scales. It allows customer feedback to be easily incorporated into software delivery.
2. Implement shift left continuous testing.
Continuous testing helps organizations focus on error prevention instead of detection. It allows testing to occur earlier in the development and release cycle. Shift left reduces the delays and code reworking that occurs when major defects are discovered late in the testing cycle. It also shortens the feedback loop and speeds up the learning process.
But beware: testing must be comprehensive. With more frequent release cycles and shorter development periods, it is important that security testing — such as application vulnerability scanning — and performance testing are built into the release cycle. Security teams should be included as part of the DevOps lifecycle.
3. Operational resilience is crucial to avoid uncertainty and disruption.
The trend of digital transformation has changed the way organizations operate. Many businesses are heavily dependent on their IT systems. Consequently, the scope of operational resilience spans people, process and IT.
Organizations should define reliable and practical steps to enhance their IT operational resilience to improve their IT organization’s ability to rapidly adapt and respond to dynamic changes, opportunities, demands, disruptions and threats that could trigger application failures.
In addition to the above, the following approaches can be adopted to avoid failure:

Learning from failures is key to prevent future issues. DevOps supports continuous learning. It instills the attitudes and activities required to effectively detect and analyze failures.
Automating code testing and the provisioning of environments will reduce time spent on those two tasks and ensure that environments are based on the same configurations. Going beyond automation, consider orchestration, which is critical in today’s cloud landscape. Orchestration pulls together automation, integration and best practices to ensure smooth and rapid delivery. Using IBM Cloud Orchestrator and Urban Code Deploy, for example, gives organizations access to ready-to-use patterns that help accelerate configuration, provisioning and deployment. It reduces IT administrator workloads and manual tasks that can lead to errors.
Decomposing your monolithic applications into microservices simplifies the application as a whole and reduces the time it takes to find and fix failures. In addition, adopting a microservices architecture helps reduce the impact of failure in one microservice. It can be isolated from the other microservices in the system. Dependency will be less of a concern and rolling back changes is much easier than with monolithic designs.

Overcoming risk to avoid failure should be a priority. The ability to use patterns to define consistent environments eliminates the failures that occur through configuration inconsistencies. Operations and development must work together to create the provisioning process so developers create their test environments the same way they are created in production.
As the ADT study the points out, organizations increasingly look to continuous deployment to recover from failure. Technologies such as IBM UrbanCode Deploy can help transform the deployment process, while practices such as shift left and DevOps will increase the feedback loop and reduce failure.
Connect with IBM Cloud Advisor Osai Osaigbovo on LinkedIn.
The post 3 ways to avoid failure in application deployment appeared first on #Cloud computing news.
Quelle: Thoughts on Cloud

The need for speed in IaaS platforms

When you look at the rapidly evolving Infrastructure-as-a-Service (IaaS) market, you might get the impression that it’s a race to the bottom with commodity hardware, inexpensive storage options, ongoing price cuts and the like.
If a cloud service provider can’t differentiate itself, things can get very ugly. That’s why one of the key tenets of IBM Cloud is delivering unprecedented speed and performance. Automating and delivering raw computing power from bare metal servers with cutting-edge GPUs (graphics processing units) can’t just be commoditized.
An example of this raw performance can be found in recent benchmark tests we conducted with our friends at MapD and Bitfusion. Together, we were able to scale up to 64 Tesla K80 GPUs across 32 servers to filter, query and aggregate a 40-billion-row dataset in just 271 milliseconds. To put that in context: together with our partners, we can scan 147 billion rows per second.

The ability to more easily pool and scale GPUs spanning multiple nodes into a single system is a significant breakthrough, maybe even a game changer. It enables businesses to manage the most complex, compute-intensive workloads — from deep learning and cognitive to big data analytics — using an affordable, on-demand computing infrastructure. The ability to explore data and run queries in near real time from the IBM Cloud gives users “supercomputer”-like performance.
Think of IBM Cloud as the Bloodhound (the project that aims to build a car that can break the world land speed record) of IaaS platforms. In many ways, IBM is trying to set its own, new cloud speed record by building the world’s fastest and most scalable cloud platform. IBM is aiming to democratize access to supercomputing resources via the cloud. Companies in industries including financial services, telecommunications, retail and social media gain a real, competitive advantage in being able to run queries like the ones we’ve tested.
Stay tuned. IBM has more HPC news coming.
The post The need for speed in IaaS platforms appeared first on #Cloud computing news.
Quelle: Thoughts on Cloud

Continuous delivery tops IT execs’ priority list

Innovation is undoubtedly all around us, from advancements in AI to quantum computing. Who wouldn’t want to capitalize on the value of digital transformation?
But what’s truly driving these big moves? It’s due in no small part to the ability of IT organizations to speed software delivery.
Enterprise Management Associates’ latest research report, derived from a survey of 600 executives conducted in October 2015, lists best practices for DevOps and continuous delivery at high-performing companies. It likewise evaluates the role of automation and release management tools in promoting digital transformation.
According to the report, businesses are indeed making the connection between accelerated delivery of software services and business growth. In fact, they are overwhelmingly making “automation of the continuous delivery process” their top technology-related initiative for supporting digital transformation this year.
This sentiment bodes well for automated release management solutions such as IBM UrbanCode Deploy, which helps companies reduce — if not completely eliminate — the potential pitfalls typically associated with the software deployment process.
Not just IT’s problem
As the momentum of innovation ramps up, IT and departments focused on business transformation are increasingly reliant on DevOps and continuous delivery.
What may be surprising is that the drivers for continuous delivery are not purely an IT problem to solve. In many cases, they are business and consumer-related, according to EMA, which has been spearheading research on these topics over the past two years.
Companies that have been able to accelerate software speed by 10 percent are more likely to double their revenue growth than companies who aren’t focused on delivery frequency, EMA Research Director Julie Craig points out in this summary video:

Craig points out:
If you aren’t able to deliver software faster, then your competition is going to continue to outpace you in terms of growth. It’s when you get automation in place that you can seamlessly deliver software releases in a way that supports speed at scale, and both speed and scale support quality.
More key findings

Ninety-seven percent of respondents have DevOps teams within their companies with dedicated personnel, 60 percent of whom are considered dedicated employees.

Companies in which DevOps interactions were rated as excellent or above average were 11.5 times more likely to have double-digit revenue growth than those who rated these interactions as average or poor.

Production troubleshooting was the biggest bottleneck to accelerating continuous delivery.

Download the 17-page report summary, “Automating for Digital Transformation: Tools-Driven DevOps and Continuous Software Delivery in the Enterprise.”
 
The post Continuous delivery tops IT execs’ priority list appeared first on #Cloud computing news.
Quelle: Thoughts on Cloud

Direct Link connection options for hybrid clouds

Companies are striving to implement more cost-effective IT environments using hybrid clouds that include both on-premises and off-premises resources. Direct Link on IBM SoftLayer, introduced in my previous blog post, helps IBM Cloud clients integrate their private, public and hybrid clouds with high performance, security-rich connections.
Here’s how Direct Link connects your IT resources.
Physical connectivity
Physical connectivity occurs through a dedicated fiber connection (one or 10 Gigabits per second) that links a customer’s service equipment and the network equipment in an IBM point of presence (PoP). The method of connection depends on the type of connection:

Direct Link NSP: Customers directly connecting their existing data centers to the IBM Cloud would terminate a telco-provided Ethernet, MPLS, DWDM or SONET connection into their service endpoint equipment within an IBM PoP, then run a fiber cross-connect from that equipment to the IBM patch panel. The panel connects to ports on the IBM cross-connect router (XCR) in the PoP, which routes traffic to an IBM data center and ultimately through IBM’s carrier-grade, private global IP network to its final destination. Customers are responsible for purchasing the physical fiber cross-connect.
Direct Link colocation: Customers with colocation facilities within the same building or campus as an existing IBM data center would terminate a redundant fiber connection into a cross-connect panel installed in the POD (server farm) where their compute resources are provisioned. Customers are responsible for purchasing the fiber cross connect from their colocated facility to the IBM data center patch panel, which can span different floors in the same building or different buildings.
Direct Link cloud exchange: Customers using an IBM-approved cloud exchange partner would terminate a telco WAN service into an IBM PoP (same as Direct Link NSP) and also into the cloud exchange partner’s software-defined network, creating secure point-to-multipoint virtual connections among their private networks, the cloud exchange and the IBM Cloud.

(See the Direct Link FAQ for more information.)

Network connectivity
Network connectivity is defined by routing policies for an organization’s IP address space. Public IP addresses are universally assigned and may be static (reserved for a particular network resource, such as a server) or dynamic (they change as resource demands change but come from an assigned public pool). Private IP addresses are used for internal traffic shielded from the public internet by network elements such as routers or firewalls. This makes it possible for the same IP address space to be used by multiple parties simultaneously without conflict.
IBM Direct Link has three network connectivity options:

Dual IP remote hosts: Customers add additional IP addresses (or reassign the IP addresses for existing on-premises, colocated or cloud exchange hosts) to include public IP address space assigned to the IBM Cloud. This allows IBM to securely route customer traffic between the customer’s private network and the IBM network.
Network address translation (NAT): Customers configure NAT on network elements acting as private network gateways (usually a router or firewall). This allows assigned public IP address space to be used on private networks without conflict, since public IPs are converted to private IPs (and vice versa) as they cross the NAT gateway. For customers with private IPs that conflict with IBM IPs, NAT can be provisioned in both directions (source and destination).
Bring-your-own IP (BYOIP): Customers bring their assigned public IP address space into the IBM private network. Customers must create generic routing encapsulation (GRE) or IP security (IPSec) tunnels between their on-premises or colocation network and the IBM network. They can then use any IP address space they choose on the private network, as long as there are no conflicts with IBM’s public or private address space, and then route traffic across the tunnel between networks.

These methods of connectivity contribute to providing the security and performance that hybrid clouds require in enterprise-scale IT environments.
To learn more about Direct Link and other features and technology available with IBM SoftLayer, check out our Cloud How-To webcast series.
The post Direct Link connection options for hybrid clouds appeared first on #Cloud computing news.
Quelle: Thoughts on Cloud