Microsoft recently announced the release of Microsoft Azure Service Fabric, a platform that supports the development of applications created using an assemblage of independent microservices as exemplified by the concept of Docker containers, for example. While Azure will be supporting Docker in a subsequent release of Windows Server, its support for microservices by means of the Microsoft Azure Service Fabric allows developers to enjoy the benefits of an architecture that uses Microsoft Azure’s indigenous microservices technology to create discrete application components that collectively enable enhanced scalability and the design of low latency, computationally intensive applications. Not surprisingly, Microsoft Azure Service Fabric features the ability to orchestrate and automate microservices in conjunction with application lifecycle management functionality for the distributed systems that are typically characteristic of applications composed of microservices. The platform also supports Visual Studio tools such as designers, editors and debuggers that facilitate the development, deployment and ongoing management of applications across a variety of operating systems, environments and devices. In the same way that Amazon Web Services recently rendered available the same machine learning and data science platform used by its own data scientists in the form of Amazon Machine Learning, the Microsoft Azure Service Fabric delivers the same core technology that Microsoft Azure has thus far used for Skype for Business, DocumentDB and Bing Cortana.
The platform’s deep experience with enterprise-grade applications that serve millions of users and files means that it “intrinsically understands the available infrastructure resources and needs of applications, enabling automatically updating, self-healing behavior that is essential to delivering highly available and durable services at hyper-scale.” As a result of its ability to understand the interplay between infrastructure, applications and distributed systems, the Microsoft Azure Service Fabric delivers a platform for development based on microservices designed to accommodate the needs of hyperscale applications. As such, Microsoft Azure Service Fabric constitutes a pre-packaged response and counterweight to the increasing traction of Docker technology by presenting Azure customers with a production-grade platform that supports stateless and stateful microservices while additionally featuring micro-services orchestration and automation, even though Microsoft plans to support Docker containers on its Azure platform in the future. In a nutshell, the Microsoft Azure Service Fabric gives developers much of the functionality of Docker and an attendant Docker management platform with the added feather in its cap that the platform has been used for years in production-grade environments for household name products such as Skype and Bing, while concurrently paving the way for streamlined usage of Docker containers on the Azure platform as well.
Cloud application security vendor Elastica recently announced a partnership with Cisco that allows Cisco to resell Elastica’s Cloud Access Security Broker solutions to its customers. The partnership between Elastica and Cisco gives Cisco customers access to Elastica’s CloudSOC portfolio which includes IT risk assessments, deep visibility regarding live cloud application traffic and the ability to detect risky actions on the part of enterprise employees. Elastica’s Shadow IT risk assessment enables customers to prioritize risk ratings of applications in addition to analytics on cloud application usage within an organization. Meanwhile, Elastica’s Shadow Data analytics provide insight into data governance and compliance-related risks with a view toward providing direction regarding how improved data governance and renewed approaches to IT compliance can mitigate an organization’s overall IT risk. Another core Elastica CloudSOC product offering provides deep insight into live cloud application traffic and user activity by leveraging machine learning and semantic technologies that can identify compromised user accounts, high risk transactions and threatening user accounts by means of a proprietary, algorithmic ThreatScore. Elastica’s cloud application security solution draws upon cloud application API data via Elastic securelets that connect to the backend of cloud applications, live cloud application data regarding networking traffic as well as usage patterns and log data from third party security solutions. Elastica’s partnership with Cisco complements solutions such as Cisco’s Cisco Cloud Web Security (CWS), Web Security Appliance (WSA) and Cisco Adaptive Security Appliance by providing deep visibility into analytics related to the security of SaaS and cloud applications more generally. Moreover, Elastica stands to gain access to Cisco’s expansive channel of customers while Cisco benefits from the data science expertise that Elastica brings to the conversation regarding cloud security. Expect to hear more about Elastica as its partnership with Cisco deepens and fuels accelerated product innovation in collaboration with the needs of its evolving customer base.
Cloud infrastructure and hosted IT solution provider CenturyLink has announced the acquisition of Portland-based Database as a Service vendor Orchestrate. Orchestrate provides a database as a service platform specifically engineered for rapid application development. Orchestrate delivers multiple databases that allow developers to “store and query JSON data” in ways that empower developers to integrate geospatial, time-series data, graphs and search queries into applications without worrying about managing the operations of the databases themselves. Ochestrate’s ability to provide a portfolio of NoSQL databases means that CenturyLink customers stand to enjoy enhanced abilities to build high performance, agile, big data applications for use cases that involve real-time streaming data, internet of things use cases and mobile applications. CenturyLink’s acquisition of Orchestrate complements its recent acquisition of predictive analytics vendor Cognilytics as well as the launch of Hyperscale, its high performance cloud instance offering designed for big data and computationally intensive workloads. The acquisition of Orchestrate illustrates the increasing confluence of cloud and Big data product offerings as vendors increasingly seek one platform to fulfill their infrastructure, application development, data storage and analytic needs.
According to a 451 Research Vendor Window assessment, the battle for IaaS leadership has intensified even though Amazon Web Services (AWS) remains the clear front-runner. Respondents to the 451 vendor evaluations revealed that 57% of enterprise customers use AWS whereas Microsoft Azure is used by 42% of customers. AWS was cited as the most important customer in 35% of all cases, ahead of Microsoft Azure, which garnered 20% of votes in the same category. Rackspace earned the highest ratings for the IaaS vendor capable of fulfilling Guaranteed SLAs and was tied with AWS for its ability to fulfill customer needs. While AWS received high ratings with respect to Experience and Technical Innovation, Microsoft Azure, in contrast, was rated lower than most of its competitors with respect to Experience and Support for Open-Source Software. Meanwhile, in the private cloud market, the 451 Research Vendor Window Assessment found that VMware claims a presence in 70% of enterprises with its ESX and vCloud virtualization platforms. Nevertheless, the survey also found that more than 70% of VMware customers have deployed other solutions for private clouds such as OpenStack or Microsoft Cloud OS, for example.
Michelle Bailey, Senior Vice President, Digital Infrastructure and Data Strategy at the 451 Research Group, remarked on the significance of the findings as follows:
While the 2015 Vendor Window for IaaS shows Amazon Web Services as the clear leader based on multiple metrics, Microsoft Azure, Rackspace and VMware’s vCloud Air are becoming competitive challengers. As more mainstream customers move business-critical workloads to cloud environments, the decision criteria for evaluating potential vendors change relative to early cloud adopters, and in turn so do the vendors under consideration.
Here, Bailey notes how the IaaS assessment reveals the emergence of “competitive challengers” to the leadership role of Amazon Web Services as the criteria for IaaS vendor selection evolves in relation to the evolving maturity of cloud adoption within the enterprise. The bottom line here is that, even though Amazon Web Services remains the most widely used and, in many ways, respected vendor within the IaaS space, enterprises are increasingly reviewing alternative options to AWS, particularly as the space features an increasing number of robust options that can variously go toe to toe with AWS regarding attributes such as customer service and ability to support SLAs. More importantly, the battle for IaaS market share is likely to become even more competitive as the progressive maturity of Big Data technologies and analytics means that enterprises are likely to seek cloud platforms that can not only support, but also streamline and simplify the adoption of Hadoop and NoSQL. Regardless, exciting times are ahead for the cloud industry as IaaS vendors mature their product and service offerings in ways that give customers the confidence to select multiple vendors to minimize risks of vendor lock-in while concomitantly enriching their knowledge of the IaaS space by sampling the heterogeneity of offerings available on the market today.
The following blog post is republished from the StackEngine blog by StackEngine CEO Bob Quillin, with permission. Cloud Computing Today had the pleasure of speaking with Bob Quillin about the contemporary interest in container technology and the concomitant pressures the industry faces with respect to container management in the context of the piece he authored below.
Container Days Austin, CoreOS, & Google
The unprecedented success of the first Container Days Austin (CDATX) and the latest series of product announcements from Google, CoreOS, and Rancher (plus soon to be others for sure) highlight two competing trends in the Container community today. Every POC (proof of concept), lab trial, and testbed evaluating Docker and containers has to fight through these issues:
- Mass Confusion: too many puzzle pieces
For most organizations, especially enterprises, there are simply too many puzzle pieces to fit together. From where to start, to which app to convert, rewrite, or start anew, which distro to run, container platform to evaluate, or service discovery to use, not to mention networking, storage, build process integration, deployment, scheduling, orchestration, and so on – the process is just too complex. The sessions at CDATX ran the gamut up and down the container stack with attendees working diligently to put the puzzle together themselve
- Intense Interest: strong belief that this puzzle is the one
Containerization arrives at a time in the market when virtualization has laid the groundwork for the next infrastructure revolution. IT, DevOps, and development teams are ready to take the next step to achieve greater capex (capital expenditure) cost savings and opex (operational expense) efficiency through automation. Furthermore, the DevOps movement has now crossed over to mainstream IT, and containers and Docker are the killer app for DevOps. Interest is beyond belief, and the senior practitioners who dominated CDATX attendance were on a clear mission to understand and learn.
It’s not surprising that the Docker market has too many jigsaw puzzle pieces to put together. This has been a developer-driven, open-source centric market built on the concept of pluggable components that developers compose together into a production application. That’s a hard problem to solve – so hard that many folks have just decided to jump to a PaaS-like approach. And if the market doesn’t start to pre-assemble the pieces for customers, we’ll likely see more and more move to a PaaS model that hides this underlying complexity from the end user. While there is nothing wrong with the PaaS model, the big opportunity for a new breed of solution combines the ease of developer deployment of a PaaS with the infrastructure control of an IaaS like Amazon AWS.
Puzzles or Portraits?
So how are we all responding to these challenges? There are several movements afoot that attack this problem:
- The Free Market Approach: create more puzzle pieces
If you don’t like the Docker platform or available Linux distros, then build your own that’s smaller, skinnier, more secure, more enterprise-ready, more performant, more whatever it is you think the ecosystem is missing. While this approach may not help Docker customers to simplify the puzzle, it’s not realistic to think that Docker can or will solve all the problems out there on its own. We’ll likely see a variety of fixes, flavors, and options offered by emerging startups and established vendors alike.
- The Top-Down Approach: pre-assemble the puzzle with your own pieces
While the whole puzzle can’t yet be assembled, vendors like CoreOS and Google (with their latest Tectonic initiative) and Docker itself see the need to begin to vertically integrate parts of the stack – to make it easier and more commercially viable. That is, if the ecosystem is becoming too fragmented (with too many puzzle pieces) then vendors may try to take the situation into their own hands and simplify. The question again here is, does the customer win?
- The Agile Market Approach: start with a picture – add pieces as you go
StackEngine firmly believes that we need better pieces to the puzzle to make everything enterprise-grade and production ready, from Docker to containers to the associated compute, networking, and storage optimizations. That push should never end. But enterprises and organizations committed to DevOps methodologies need to apply agile methods throughout. They require an end-to-end solution where they can pay less attention to assembling the puzzle pieces and more to cutting licensing costs, deploying apps more reliably, quickly and frequently, with the freedom to plug-in new jigsaw puzzle pieces as they go and as they choose.
If you want to be part of the discussion on where and how this industry will move forward, keep an eye out for the next Container Days unconference, where you can come ready to learn more about this emerging puzzle and/or lead the conversation. Container Days is coming to Boston with Container Days Boston on June 5-6 and plans are in the works to head to Silicon Valley next. They are just getting organized and are looking for sponsors. Also, more cities are in the works and let us know if we can help get things off the ground! Finally, Container Days Austin 2016 is already in planning mode so check out Docker Austin in the meantime to get plugged in.
On April 15, Logentries announced an AWS CloudTrail log auditing service that aggregates log data produced by the Logentries Data Insights Service into a weekly report that provides insight into an AWS infrastructure. In filtering, aggregating and delivering data visualizations of log data, the CloudTrail log auditing report facilitates access to actionable business intelligence regarding log data by taking the onus of log data analytics away from the customer. Log data in the CloudTrail report features metrics related to the most active users, common security-related actions, unauthorized access, system activity over time and a categorization of AWS errors with metadata on their frequency. Logentries CEO Trevor Parsons elaborated on the significance of analytics on log data by noting that the “signal to noise ratio” for large volumes of log data can be low, meaning that the amount of meaningful log data within a particular log dataset is often disproportionately small in comparison to the larger volume of log data. Because of the way in which the most significant log data is often buried within a haystack of log data that effectively constitutes “noise,” the Logentries Data Insights Service is all the more useful because it gives customers enhanced access to the most salient insights from their log data. Available in May, the free AWS CloudTrail log auditing service can be accessed by Logentries customers with AWS deployments by signing up for the AWS CloudTrail weekly report via email.