Condusiv Unveils Expansive I/O Performance Study

Condusiv Technologies Corp., in software-only storage performance solutions for virtual and physical server environments, announced results from one of the largest ever gathering of performance data from 3,450 systems at over 100 sites that evaluated V-locity and shared their before/after data with Condusiv servers.

The unfiltered data was provided by virtualized organizations using Condusiv’s newest benefit analyzer tool embedded within V-locity to gather real-time performance data from VM to storage, before and after deploying V-locity. This expansive I/O performance study reveals never-before-seen data about the extent of I/O inefficiencies in virtual environments and the impact V-locity I/O reduction software has on production workloads when used to solve root cause performance issues. As examples, more than half of all systems saw a minimum of 50% decrease in read I/O from VM to storage; nearly half of all systems saw a 50% jump in throughput and all systems experienced an average of 40% reduction in I/O response time.

Condusiv’s latest benefit analyzer tool collects workload data across a range of metrics including throughput, latency, IO/s, I/Os per gigabyte, I/O density, and more, which helps administrators get a complete before/after snapshot of the performance benefits provided by V-locity in their real-world production environment before making a purchase commitment.

Customers Validate Value of V-locity
“Once we installed V-locity, we monitored our IO/s and saw a near 50% decrease in I/O from VM to storage according to our daily averages and peaks across our heaviest SQL-based workloads,” said Jamie Rabenstein, IT manager, Victor Community Support Services.

“Facing a $2 million storage purchase to solve our performance issues didn’t sit well with any of us. We heard what V-locity I/O reduction software had done to help other Meditech hospitals, so we were eager to try it for ourselves. After deploying V-locity and conducting a before/after performance analysis, we found that V-locity doubled the speed of our patient records. V-locity provided enough speed to meet our application SLAs, so there was no longer a need to make a $2 million storage purchase for increased performance,” said Tom Swearingen, manager infrastructure services, Christus Health.

“When we first tried V-locity in our virtual infrastructure, we didn’t know how much of our performance was being penalized by the ‘I/O blender’ effect at the hypervisor and the surplus of split I/Os coming from the Windows VM. As it turns out, our performance was being severely dampened. It wasn’t until V-locity optimized our I/O streams that we realized some of our systems were processing half as many MB/s as they should from VM to storage. Since we’re no longer dealing with so many small, split I/Os and random I/O streams, V-locity has enabled our CRM and ERP systems to process twice the amount of data in the same amount of time. The best part is that we didn’t have to spend a single dime on expensive new hardware to get that performance,” said Kevin Schmidt, director of IT, I.B.I.S, Inc.

Key Metrics Evaluated by the Performance Benefits Analyzer Include:

• Percent reduction of total I/Os going to storage
• Increase in throughput
• Reduction in average I/O response time
• Increase in total number of IO/s
• Average size of I/O increase
  

In addition to the reduction in percentage of total I/Os going to storage, several other categories were reviewed for performance gains from the ‘before’ and ‘after’ data, resulting in these additional key findings from the analysis:

• Reduced read I/O to storage: ESG Lab calculated 55% of systems saw a reduction of 50% in the number of read I/Os that get serviced by the underlying storage, and 27% of systems saw a 90% or more reduction in read I/Os.
• Reduced write I/O to storage: As a result of I/O density increases, ESG Lab witnessed a 33% reduction in write I/Os across 27% of the systems. In addition, 14% of systems experienced a 50% or greater reduction in write I/O from VM to storage.
• Increased throughput: ESG Lab witnessed throughput performance improvements of 50% or more for 43% of systems, while 29% of systems experienced a 100% increase in throughput, and as much as 300% increased levels of throughput for 8% of systems.
• Decreased I/O response time: ESG Lab calculated the total time required to process all requests for each system and concluded that on average, systems with 3GB of available DRAM achieved a 40% reduction in response time across all I/O operations.
• Increased IOPS from DRAM: ESG Lab found that 25% of systems saw IOPS increase by 50%, and a small group of 25 systems achieved a 1000% IOPS improvement.

“The key take-away from this analysis is demonstrating the extent of I/O inefficiencies that plague virtual environments by robbing up to half of available I/O bandwidth from VM to storage due to I/O characteristics that are much smaller, more fractured, and more random than they need to be,” said Brian Morin, SVP, global marketing, Condusiv. “These inefficiencies are akin to pouring molasses on systems, and it isn’t until someone evaluates V-locity to solve these root cause performance issues that they get the ‘ah ha’ moment of exactly how much performance was being robbed and how easily these issues can be corrected.“