Watching VMware roll out vSphere was a bit surreal. VMware did a great job of explaining the mission of vSphere 5 (attack the next 60% – virtualize any application). VMware did a great job of positioning its cloud suite (vSphere 5 and all of the products announced along with it) as the suite of functionality needed to address that next 60% of un-virtualized workloads. In fact all of the analyst briefings that occurred before the actual launch focused upon these two elements as did the first 90% of the launch. And then… Continue reading Could the new vSphere vRAM Pricing Slow Down Virtualizing Business Critical Applications?
VMware’s vSphere team has done it again. The most important, and best systems software company on the planet is again announcing a major upgrade to its platform that once again raises the level of its came into a different orbit than the pretenders and the contenders. Continue reading vSphere 5 – Virtualize Business Critical Applications with Confidence
ExtraHop is a “network aware APM” vendor that has historically focused on putting an appliance on a mirror or span port of a switch and then using deep packet inspection techniques and TCP/IP flow analysis techniques to understand the performance of applications running on the network from the perspective of the network. Today, ExtraHop has announced the addition of a virtual appliance to its product line – allowing for the data collection appliance to be implemented entirely in software, and to us the virtual mirror port on the vSphere vSwitch to collect its data.
Adding a virtual appliance to the product line allows ExtraHop to accomplish two very important new things. The first is that conversations between VM’s on one host are now visible, and their response time can now be part of the picture of the response time of the entire application. But the second is far more important. Since ExtraHop can already measure the true performance (end-to-end response time) of every application on a physical TCP/IP network, adding the virtual appliance means that now ExtraHop can cover both the physical network (and applications that are not virtualized yet), as well as the virtual network (covering applications that are either completely or partially virtualized).
The New p2V Process
It is very clear that the difficulty of virtualizing business critical applications for many enterprises means that for business critical and performance critical applications the P2V process is either broken or completely fails to exist in a relevant manner. Measuring how much CPU and memory and application uses is a necessary but a completely insufficient step towards virtualizing a business critical application.
What is required to virtualize a business critical and performance critical application is a P2V process that is focused upon the performance of the application in the eyes of its owner and its owners. That means that performance is equal to response time, not resource utilization. This process must therefore start with building a response time and load profile of the application in its physical home, agreeing on the aspects of that profile that will be the basis of the SLA when that application gets virtualized, and then continuously updating that profile as the application gets virtualized.
Criteria for Application Performance Based P2V
With this announcement ExtraHop joins the fray for the set of solutions that are viable candidates for being the basis of an applications response time based P2V process. The criteria for selecting such a tool are:
- Ability to measure response time for every application in the environment (or at the minimum all X86 applications that use TCP/IP).
- Ability to automatically discover the applications in the environment and their topology
- Ability to automatically measure response time for each application end-to-end and hop-by-hop
- Measure response time on a continuous, real-time, and deterministic basis. This means no averages, no sampling of data, no synthetic transactions. It means seeing the response time for every transaction over the network, and being able to pass the slow ones up to the management system in real time.
- The solution should work in the existing physical environment of the application, in the new virtual envionment of the application and any mixture of the two.
- The diagnostics in such a solution should focus on providing pointers to where in the virtual and physical infrastructure the problem likely lies, as opposed to pointing to code issues which is what developer focused tools do.
Other than ExtraHop there are two other solutions that are appropriate IT Operations oriented (as opposed to developer oriented) solutions. Those two solutions come from BlueStripe and AppFirst. All three of these solutions are profiled in an upcoming post about the use of these solutions in the context of the new capabilities of VMware vSphere 5.0.
ExtraHop has now made an important contribution to the question of how to measure applications performance across physical and virtual environments. Properly deployed ExtraHop can play a critical role in helping enterprises virtualize the 60% of the remaining applications that are “hard”, “performance critical”, and “business critical”. As vSphere 5.0 is right around the corner, the timing could not be better.
ExtraHop Creates Fastest Virtual Application Performance Management (APM) Solution with New EH1000v Appliance
ExtraHop’s New Virtual Appliance Provides Ideal APM Solution for Branch-Office Deployments and Virtual and Cloud Environments
July 10, 2011 – ExtraHop Networks, a leading provider of network-based application performance management (APM) solutions, today at Cisco Live announced its virtual appliance for APM, the EH1000v. The new appliance furthers ExtraHop’s commitment to building the fastest APM solutions available by providing real-time transaction analysis at up to 1Gbps speeds for applications running in virtual and cloud environments—more than 10 times faster than competing products. As a virtualized version of the company’s passive, network-based ExtraHop Application Delivery Assurance system, the EH1000v delivers a small-footprint solution for branch-office deployments and offers deep, end-to-end application visibility for virtualized environments.
The EH1000v virtual appliance lowers IT infrastructure requirements for monitoring network and application performance across geographically distributed environments. In line with Gartner’s prediction that 20 percent of companies will own no IT hardware assets by next year, the EH1000v eliminates the need to ship and install hardware for each geographic location. With more companies using virtualized systems in remote-office applications, the new virtual appliance from ExtraHop Networks will help these organizations deploy and remotely administer a powerful monitoring solution for branch-office deployments that also can monitor traffic within virtualized environments.
As they have increasingly virtualized their systems, many organizations have lost visibility into transactions passing between virtual machines on the same physical host, leading to a growing blind spot for heavily virtualized organizations. According to an InformationWeek survey, nearly 50 percent of companies using APM solutions run more than half of their applications in virtualized environments, creating a significant amount of traffic that enterprises cannot see on the network. The EH1000v can monitor traffic that passes only through virtual switches and restore deep visibility into applications running in such environments, even those with large traffic volumes among multiple virtual machines.
“While server virtualization offers many benefits such as flexibility, scalability, and lower cost, it also can limit visibility into virtualized environments,” said Jesse Rothstein, CEO and co-founder, ExtraHop Networks. “In these cases, the physical limitations of many APM solutions present obstacles to monitoring application performance. The EH1000v gives companies a small-footprint solution that helps them to get around these roadblocks and realize the benefits of virtual and cloud technologies without compromising application performance.”
About ExtraHop Networks
ExtraHop Networks is a leading provider of network-based application performance management (APM) solutions. The ExtraHop Application Delivery Assurance system performs the fastest and deepest analysis in the industry, achieving real-time transaction monitoring at speeds up to 10Gbps in a single appliance and application-level visibility with no agents, configuration, or overhead. The ExtraHop system quickly auto-discovers and auto-classifies applications and devices, delivering immediate value out of the box. ExtraHop Networks provides award-winning solutions to companies across a wide range of industries, including ecommerce, communications, and financial services. The privately held company was founded in 2007 by Jesse Rothstein and Raja Mukerji, engineering veterans from F5 Networks and architects of the BIG-IP v9 product.
You teenager takes the car out for a drive. Your teenager does an hour of driving and comes back with an additional 30 miles on the odometer. Your teenager also comes back with a speeding ticket for doing 60 MPH in a 30 MPH zone. When confronted with this unacceptable behavior, your teenager responds that their average speed while driving the car was only 30 MPH and that the police officer was “unfair” for focusing only on the peak and not the average. You explain the obvious – that you only have to go 60 MPH in a 30 MPH zone for an instant to be guilty. Continue reading Virtualization Performance Management and Misleading Averages
Back in May, VMTurbo broke some new ground in terms of what was available in terms of free vSphere monitoring tools, by making some right-sizing reports that were previously part of paid for capacity management and planning solutions available for free. Now VMTurbo has built upon the earlier breakthrough by making a fully functional performance and capacity management solution that is not time limited nor limited to size of environment available in the form of the VMTurbo Community Edition (Download).
The What and the Why
VMTurbo, like any software company, is in business to make money. So of course there are paid for editions of the product. The table below shows the differences between the free edition (the Community Edition) and the two paid for versions.
The key thing to realize about VMTurbo is that unlike many vendors who simply collect vCenter data, store it, analyze it, dashboard it, report on it and alert on it, VMTurbo has some very unique technical capabilities (which of course they charge for in their paid products), that allow VMTurbo to make what is rapidly becoming “commodity” vSphere monitoring functionality available for free.
The secret sauce at VMTurbo is the ability to price (in virtual dollars) scarce resources, to allow you to assign budgets to workloads (applications running in VM’s), and to then either make recommendations or to automatically execute those recommendations. Strategically VMTurbo is therefore further down the path towards Service Assurance (the idea that the infrastructure adjusts automatically to ensure that the most critical applications perform as required) than any other vendor. The only thing missing from VMTurbo’s bag of tricks is a true view into applications performance (response time) which will doubtless be handled via partnerships with various APM vendors.
These unique capabilities (and the ability to charge for them in the paid products) allows VMTurbo to make baseline performance and capacity management for vSphere environments available for free. This represents a significant redefinition of where the value is and where the value is not in the virtualization performance management business.
We have long believed that virtualization, IT as a Service, Private Clouds, and Public Clouds represent tectonic shifts in the systems management market. These shifts affect every segment of this market, including configuration management, provisioning and updating, infrastructure availability management, capacity management and of course applications performance management.
Applications Performance Management is destined to assume a key role in the future IT environment that is characterized by workloads getting dynamically provisioned (through IT as a Service or Private Cloud management offerings), being subject to dynamic operations (through vMotions triggered either manually, or automatically by DRS or other service assurance solutions), and getting distributed across physical data centers either within an enterprise, or across a private cloud and a public cloud.
These new ways in which applications will get deployed create new requirements for how they need to be managed, hence the central role of APM solutions in management. In the physical world, very few applications were actually managed by APM solutions. In the virtual and private cloud world, we assert that every application needs to be managed for response time, and that next generation APM solutions are what should do the managing. In order for APM solutions to assume this dramatically more prominent role, they need to evolve in the following ways:
- First generation APM tools have been notoriously expensive to purchase, and difficult to configure and maintain. The new generation of APM tools work on a “zero-config” basis. This means that they can automatically discover the applications in the environment, automatically discover the topology of the applications in the environment, automatically configure themselves to time the transactions of interest, and then automatically instantiate and maintain the proper level of monitoring.
- First generation APM tools assumed a fixed and static infrastructure for the application where the application support team knew at all times on which servers which components of the application were running and knew that those servers were inside of the four walls of their data center. The new dynamic environment is characterized by things moving around (between data center, or between organizations, i.e, to a pubic cloud on demand). Next generation APM solutions now need to automatically work when applications are dynamically scaled out and/or distributed in this manner.
- In the past the performance of many applications was simply inferred by looking at the resource utilization profile of the application. In this new environment that approach breaks, and the only way to know what the performance of the application really is, is to constantly measure the response time that the application is delivering to end users.
- As applications become mashup’s of internally developed services and SaaS delivered applications it will become increasingly difficult to know what the experience is of the end user by looking at the application system (as SalesForce.com is not going to allow you to install your APM solution on their servers). Therefore an understanding of the actual experience of the end user will become of increasing importance.
There are several vendors who are leading the charge in terms of redefining the APM space to meet these new criteria. These vendors are:
- New Relic – who offers a hosted (Monitoring as a Service) APM solution that supports Java, .NET, Ruby, and PHP applications and that includes both APM and Real User Monitoring (RUM) functionality.
- AppDynamics – who offers both an on-premise and hosted solution that targets scaled out open source deployments of primarily Java based applications – and that does a great job of automatic transaction discovery.
- BlueStripe – who offers a solution that works for any application that runs on Windows, Linux, AIX, or SunOS, and that discovers those applications, discovers their topology, and determines end-to-end and hop-by-hop response time.
- dynaTrace – who has made their mark by using their Purepath technology to be the only vendor that can trace individual transactions from the end user’s browser through the web servers, the middle tier servers, the database servers, through various layers of enterprise middleware, and back to the end user.
The news today is that Compuware has acquired dynaTrace. This is an extremely significant development as this is the first example of one of the “traditional” systems management vendors (IBM, HP, CA, BMC, Compuware, Quest), stepping up their game to acquire one of these new innovators that address these new use cases and requirements.
The full acquisition announcement is copied in below.
Compuware Acquires dynaTrace software, Setting the New Standard for Application Performance Management
Meets Growing Business Demand for IT Agility, Enables Rapid Delivery of High-performing Modern Apps and Cloud-based Services
DETROIT–July 6, 2011–Compuware Corporation (NASDAQ:CPWR), the technology performance company, today announced that it has acquired privately held dynaTrace software. The $256 million cash acquisition closed on July 1, 2011.
“Organizations today depend on the rapid development and delivery of high-performing applications to drive revenues, customer satisfaction and brand,” said Compuware Chief Executive Officer Bob Paul. “To meet these demands effectively, IT organizations must have visibility into the performance of every transaction, from development, through test and in production. Together, Compuware and dynaTrace APM solutions allow IT to meet business demands for performance and agility through unbeatable insight into the user experience – whether in cloud, complex or traditional environments.”
Headquartered in suburban Boston, dynaTrace employs 180 people around the world. Substantially all of these employees, including the leadership team, are expected to remain with Compuware. dynaTrace has a 92-percent win rate and a five-year CAGR of 125 percent because its unique PurePath technology solves the multi-billion dollar problem of optimizing application performance in a different and better way.
“The APM market is undergoing a transformation; old tools simply can’t meet the competitive requirements of modern applications,” said John Van Siclen, Chief Executive Officer of dynaTrace software. “Compuware sees the same market shift and is committed to building next-generation APM solutions that bring customers greater value, with less effort, faster than ever before. We’re excited to be a part of the Compuware team, which will allow us to meet and increase the skyrocketing demand for dynaTrace solutions.”
dynaTrace has trailing twelve month revenues of $26 million and will add between $35 and $45 million to Compuware’s FY ’12 revenues. In FY ’13, Compuware expects it to add nearly $100 million to the top line.
Due to purchase accounting, the acquisition will have a three cents per share negative impact on Compuware’s expected FY ’12 EPS. In FY ’13, the impact is expected to be at least four cents per share positive. These estimates reflect the long-term value of Compuware’s investment strategy to become best-in-the-world in the APM market.
For complete details about how this agreement raises the bar for optimizing application performance, visit http://www.compuware.com/dynatrace.
dynaTrace was founded by Bernd Greifeneder, Sok-kheng Tang and Hubert Gerstmeyer in 2005. Ben Nye at Bain Capital Ventures invested in dynaTrace at its founding, and Salil Deshpande of Bay Partners joined the company’s board in 2008. Pacific Crest Securities served as the advisor to dynaTrace on this transaction.
About dynaTrace software Inc.
dynaTrace is the new leader in application performance management (APM). With its patented PurePath technology, the company offers the only continuous APM system on the market, transforming how applications are built, tested and managed. Hundreds of companies including Zappos, SAS, Macy’s, BBVA and Thomson Reuters rely on dynaTrace to drive better business results by optimizing performance, accelerating time-to-market for new releases, reducing application management costs, and bringing business and IT closer together. Visit dynaTrace online at www.dynatrace.com.
Compuware Corporation, the technology performance company, provides software, experts and best practices to ensure technology works well and delivers value. Compuware solutions make the world’s most important technologies perform at their best for leading organizations worldwide, including 46 of the top 50 Fortune 500 companies and 12 of the top 20 most visited U.S. web sites. Learn more at: http://www.compuware.com.