How Many Eyeballs Does It Take?  New Network Performance Insight with Thousand Eyes

(Excerpt from original post on the Taneja Group News Blog)

How many eyeballs does it take to really see into and across your whole network – internal and external?  Apparently a “thousand” according to Thousand Eyes, who is coming out of stealth mode today with a very promising new approach to network performance management. We got to talk recently with Mohit Lad, CEO of Thousand Eyes, who gave us a fascinating preview. Thousand Eyes already has a very strong list of logos signed up, and we think that they will simply explode across the market as one of the few strong solutions that really helps unify management across clouds.

One of the issues with network side performance management these days is that much of the network that IT enterprises rely on is external. Over the years there have been a few attempts at quantifying and monitoring external network issues – who hasn’t scripted some traceroute commands to check on some questionable services?  You might have many network tools  – maybe you run Netscout in-house but rely on some APM tools to monitor a SaaS provider.  Still you are likely missing all the network traversal in between, and it’s almost impossible to analyze for actual contention and hotspots in a deep topology (there is an echo here in what NetApp Balance did for layers of server/storage virtualization).

Companies like Gomez/Compuware, BMC, Keynote, and others have created vast endpoint armies (the good botnets I suppose) to test public facing web apps from many, many places at once. But if there is a problem, then what?  There is now a growing challenge to find and isolate devious performance issues hidden in complex networks hosting all kinds of internal and external apps and services that span private and public clouds, CDNs, and even DDOS protection vendors. Network topologies are just darned complex and often your visibility is extremely limited.

This is where Thousand Eyes comes in. They do offer a similar HTTP page component loading analysis like Compuware and Keynote, but the exciting thing is what and how they link that application performance with network performance constraints on end-to-end network topology mappings, regardless of where those networks traverse.  Imagine being able to tell your service provider that they are dropping packets at IP xxx in their network, and its affecting your apps x, y, and z in location a and b!

It’s a bit hard to describe in text – a picture here would really be worth thousands of words – but when there is a network issue, Thousand Eye’s uniquely focusing topology visualization (supported by “deep path analysis”) nails down the bottleneck fast regardless of if the problem is in-house, or at your ISP, service provider, CDN, etc.  And since its SaaS based, you can easily “share” your dynamic view of the problem with the support teams at those providers for quick collaborative resolution (and I expect providing a motivation for those folks to also subscribe to Thousand Eyes).

This may be one of those things that you need to see to appreciate, but I was quickly impressed by the thoughtful visual analysis of some issues that would otherwise be pretty opaque and hairy to figure out.  If you manage application performance across wide areas and multiple providers, or are responsible for network performance on complex topologies, you’ll want to check these guys out.  Network pros will probably get more out of it immediately than others, but even a server or storage guy can tell that a red circle in a service providers IP address is something to call them about.

…(read the full post)

Is Virtualization Stalled On Performance?

An IT industry analyst article published by Virtualization Review.

Virtualization and cloud architectures are driving great efficiency and agility gains across wide swaths of the data center, but they can also make it harder to deliver consistent performance to critical applications. Let’s look at some solutions.

article_virtualization-stalled-performanceOne of the hardest challenges for an IT provider today is to guarantee a specific level of “response-time” performance to applications. Performance is absolutely mission-critical for many business applications, which has often led to expensively over-provisioned and dedicated infrastructures. Unfortunately broad technology evolutions like virtualization and cloud architectures that are driving great efficiency and agility gains across wide swaths of the data center can actually make it harder to deliver consistent performance to critical applications.

For example, solutions like VMware vSphere and Microsoft Hyper-V have been a godsend to overflowing data centers full of under-utilized servers by enabling such high levels of consolidation that it has saved companies, empowered new paradigms (i.e. cloud), and positively impacted our environment. Yet large virtualization projects tend to stall when it comes time to host performance-sensitive applications. Currently, the dynamic infrastructures of these x86 server virtualization technologies don’t provide a simple way for applications to allocate a “real performance level” in the same manner as they easily allocate a given capacity of virtual resource (e.g. CPU, disk space). In addition, virtualization solutions can introduce extra challenges by hiding resource contention, sharing resources dynamically, and optimizing for greatest utilization.

The good news is that additional performance management can help IT virtualize applications that require guaranteed performance, identify when performance gets out of whack, and rapidly uncover where contention and bottlenecks might be hiding

…(read the complete as-published article there)