Rampant I/O Demands Crippling Storage Performance
Organizations struggle to gain the full lifecycle from their backend storage due to I/O growth
September 07, 2017

Brian Morin
Condusiv Technologies

Share this

When New York's Triborough Bridge opened in 1936, it was widely viewed as the end of traffic congestion. And it was – for several months. But its architect soon realized that better roadways and better bridges inevitably led to more traffic on the road. It's no different in IT.

The Input/Output Operations per Second (I/O) capabilities of modern computer systems are truly a modern wonder. Yet no matter how powerful the processors, no matter how many cores, how perfectly formed the bus architecture, or how many flash modules are added, somehow it never seems to be enough. While existing applications flourish on the latest platforms, software designers are quick to design new ones to take advantage of all this new potential. Result: just like New York City – traffic congestion in IT systems becomes a fact of life.

This is confirmed by a newly released survey by Condusiv Technologies, the third annual I/O Performance Survey, which consulted over 1,400 IT professionals and revealed some startling facts. Organizations have been keen to adopt the latest in all-flash arrays, hybrid arrays, hyperconverged architectures, PCI-e flash cards, and servers with even more cores as a solution to their performance woes. Yet a full 27 percent continue to receive user complaints about sluggish performance on mission-critical applications such as MS-SQL. They see little option but to add yet more expensive hardware to alleviate the many bottlenecks they are dealing with.

The bulk of these organizations are operating heavily virtualized Windows environments. Almost half of them admit serious problems supporting one or two of their most demanding applications. The specifics vary from business to business, but encompass the likes of MS-SQL, SAP, Oracle, Microsoft Exchange, various ERP tools, CRM, databases, BI, analytics, VMware, VDI, Splunk, financials and security applications.

A full 30 percent of those surveyed believe that the growth of I/O from applications has outpaced the useful lifecycle they expected from their underlying storage architecture. Only 41 percent consider that they are able to cost-effectively keep up with the growth of I/O. The rest face a grim future of budget restrictions in the face of urgent demands to add more server and storage hardware.

They find themselves with some tough choices. Should they grit their teeth and endure these performance issues until the next budget cycle, and hope there is some leeway in next year's budget to purchase new hardware? Or should they divert money away from other urgent IT initiatives to solve the incessant flood of user complaints?

While the latest and greatest hardware is always going to make a difference, it doesn't solve the problem in the long run. As the architects of our road systems have found, newer, wider and better roads only lead to more traffic and eventual gridlock. That's why they are turning to software as a key part of their vision for a better future. They are working hard on initiatives such as driverless cars, traffic flow optimization software, and vast road sensor networks feeding data into to traffic flow analytics systems. Certainly, such systems must be supported by new hardware. But software is being turned to as the ultimate solution to city congestion.

Similarly, in the world of ever rising I/O demands, software solutions are emerging that provide relief from performance bottlenecks. The various approaches to I/O streamlining and reduction include caching, micro-tiering, fragmentation prevention and performance tuning. They offer a way to better performance now without breaking the bank.

Brian Morin is SVP Global Marketing at Condusiv Technologies
Share this

The Latest

December 18, 2024

Industry experts offer predictions on how NetOps, Network Performance Management, Network Observability and related technologies will evolve and impact business in 2025 ...

December 17, 2024

In APMdigest's 2025 Predictions Series, industry experts offer predictions on how Observability and related technologies will evolve and impact business in 2025. Part 6 covers cloud, the edge and IT outages ...

December 16, 2024

In APMdigest's 2025 Predictions Series, industry experts offer predictions on how Observability and related technologies will evolve and impact business in 2025. Part 5 covers user experience, Digital Experience Management (DEM) and the hybrid workforce ...

December 12, 2024

In APMdigest's 2025 Predictions Series, industry experts offer predictions on how Observability and related technologies will evolve and impact business in 2025. Part 4 covers logs and Observability data ...

December 11, 2024

In APMdigest's 2025 Predictions Series, industry experts offer predictions on how Observability and related technologies will evolve and impact business in 2025. Part 3 covers OpenTelemetry, DevOps and more ...

December 10, 2024

In APMdigest's 2025 Predictions Series, industry experts offer predictions on how Observability and related technologies will evolve and impact business in 2025. Part 2 covers AI's impact on Observability, including AI Observability, AI-Powered Observability and AIOps ...

December 09, 2024

The Holiday Season means it is time for APMdigest's annual list of predictions, covering IT performance topics. Industry experts — from analysts and consultants to the top vendors — offer thoughtful, insightful, and often controversial predictions on how Observability, APM, AIOps and related technologies will evolve and impact business in 2025 ...

December 05, 2024
Generative AI represents more than just a technological advancement; it's a transformative shift in how businesses operate. Companies are beginning to tap into its ability to enhance processes, innovate products and improve customer experiences. According to a new IDC InfoBrief sponsored by Endava, 60% of CEOs globally highlight deploying AI, including generative AI, as their top modernization priority to support digital business ambitions over the next two years ...
December 04, 2024

Technology leaders will invest in AI-driven customer experience (CX) strategies in the year ahead as they build more dynamic, relevant and meaningful connections with their target audiences ... As AI shifts the CX paradigm from reactive to proactive, tech leaders and their teams will embrace these five AI-driven strategies that will improve customer support and cybersecurity while providing smoother, more reliable service offerings ...

December 03, 2024

We're at a critical inflection point in the data landscape. In our recent survey of executive leaders in the data space — The State of Data Observability in 2024 — we found that while 92% of organizations now consider data reliability core to their strategy, most still struggle with fundamental visibility challenges ...