VirtualWisdom: A flash of clarity

FLASH IS INDEED FAST, but what is often overlooked is the reality of today’s heterogeneous IT infrastructure; from a cost vs. performance perspective, spending thousands on flash arrays will not guarantee optimum performance. When it comes to the entire SAN, even with the installation of super-fast new flash arrays, overall output can be still offset by sluggish bottlenecks in other areas. Without a deep understanding of what is really going on within the unique complexity of your data center’s health, how would you know that the spend on purchasing flash or any other performance enhancing application is even the right decision?

The benefits of flash memory storage are compelling. The technology is able to deliver a far superior cost-to-performance ratio than traditional disk storage. You might imagine then that flash should be flying off the shelves and perhaps it will be in the near future. But for now the industry has not advanced to the magical inflection point where the cost delta (£s per GB) between silicon and disk ceases to exist and there are still some barriers to widespread adoption.

Leaving aside the prohibitive cost argument, there a numerous placement and packaging options which can be overwhelming and confusing. Will a dedicated array deliver a noticeable performance gain over a server-based memory extension product? Do you opt for a hardware-centric or purpose-built software-optimized design? Should you consider a SAN caching appliance or would a hybrid model (SSDs using auto-tiering in your legacy disk array) suffice?

There are endless variables that come into play when evaluating flash memory storage such as access patterns, I/O-mix, I/O parameters and load priorities that need to be carefully addressed, particularly in a multi-tenant scenario. To add to this, there is the nuisance that different vendor solutions will inevitably deliver varying real-world performance.

Any IT manager will want to be sure of what to expect from a vendor. Unfortunately the difference between what vendors’ spec sheets state (millions of IOPS /sub-millisecond latency) and what you are likely to achieve during real application testing can vary hugely. This creates a highly unattractive situation for those seeking out a reliable IT solution, but is no surprise when considering the number of variables at play.

Determining the variables
Application acceleration is certainly an appealing proposition. Indeed it’s sufficiently enticing for numerous organisations to deploy solid-state storage when often the decision-makers do not fully understand all of these factors. Often, little consideration goes into evaluating and determining how the surrounding infrastructure will respond to the introduction of memory-based storage.
When evaluating the speed of a SAN, we need to be mindful of the elements and devices that sit in the data path, as these collectively determine its overall performance. If response times are throttle-stopped by the weakest link in the transaction flow for example, then surely it makes sense to analyse storage traffic throughout the entire stack.

This may sound obvious, but without an advanced level of visibility, hundreds of thousands of pounds can be invested in shiny new hardware, based largely on a high level interpretation of I/O-intensity, which is inferred and imprecise. The underlying problem here is that almost all standard benchmarking for flash storage focusses on database trace file analysis (performance monitoring output), which looks solely at database I/O processing tasks, rather than fully characterising the workflow end-to-end. Flash vendors use this to review CPU utilization, throughput, queue depths and I/O-wait latency to highlight potential areas for improvement. While this reveals if there’s an overall path latency problem, it can’t pinpoint where and why, or propose suitable remediation.
Without a truly holistic and correlated view, there is no way of determining whether there will be potential gremlins lurking in the shadows until it is too late and you have already parted with your cash. Only then do unfortunate resource conflicts and ‘misalignments’ elsewhere in the infrastructure start to surface; wrongly configured parallelisation settings, O/S versions that aren’t optimized for flash storage, under-sized I/O interfaces, network congestion and other stressed components which are now overloaded.

One virtual Instruments customer in the UK explained how he was asked to buy and implement more than four hundred thousands pounds worth of flash arrays to support a critical application. When the application was benchmarked, using VirtualWisdom, it showed that the performance inhibitor was not the array and the investment in Flash would not have made any significant difference . This is not an uncommon experience.

It is beyond doubt that solid-state storage is a fantastic enabling technology. But all too often when it is deployed, it compounds load concentration, exposing weaknesses and bottlenecks elsewhere in the SAN, which would otherwise remain masked and undetermined.
Overcoming uncertainty
Virtual Instruments’ unique method of instrumenting the SAN, coupled with powerful analytics, means impact analysis is measured and precise, allowing customers to set an application performance SLA or guarantee back to the business It’s real-time acquisition and correlation of performance data, extracting every individual read/write command in the data-flow directly from the Fibre Channel protocol, provides an unbiased view of the entire SAN and explicit, dynamic measurement of all interactions between the layers, which could cause application latency.

VirtualWisdom gathers hundreds of critical performance metrics from various sources at the physical layer (something that non-physical monitoring tools simply can’t do), providing full transaction tracing, which can be used for performance impact modelling and evidence-based decision support, completely removing the guesswork, minimising risk and ensuring the success of your flash implementation.

The VirtualWisdom platform helps to eliminate performance ‘blind spots’ and to prevent configuration-based issues from impacting the performance and availability of your flash investment. Moreover, it ensures the infrastructure is properly balanced and optimally tuned, proactively resolving problems before they occur, rather than simply shifting the bottleneck elsewhere.
In the face of continual change and increasing complexity (layers of abstraction & the dynamic demands of a virtualized compute environment), today’s IT manager needs the tools at hand to make informed, hard-headed decisions. This requires detailed planning and consideration, as enterprises strive to virtualise more latency-sensitive applications.

An assessment should be a pre-cursor to any flash storage implementation, to verify the likely impact, validate ROI and TCO computations and provide valuable prescriptive guidance for optimisation and forward-engineering of the SAN.