refresh(PIX) - Fotolia
Any slowdown in the pace that technology gets faster through Moore’s Law would lead to increases in the price of adding chips and processing power in datacentres, a study from Futuriom has warned.
Although 37% of the 218 IT professionals who took part in Futuriom’s Untold secrets of the efficient data center study for Mellanox, expect Moore’s Law to continue, almost a quarter believe it will decelerate.
The study warned that if there is a slowdown in Moore’s Law, the industry would need to find alternative ways to optimise datacentre performance.
When respondents were asked which aspect of hyperscale cloud operations they would most like to emulate in IT operations, highly efficient utilisation of servers and storage topped the list. Futuriom’s analysis said the results show that datacentre managers are trying to avoid deploying more server and storage hardware.
Almost one-fifth of the datacentre managers surveyed said they wanted to emulate the flexible, converged ethernet networking used by hyperscale cloud providers. Automated infrastructure deployment, management and monitoring (17%) and simplified resource provisioning, reporting and billing (15%) were also seen as desirable properties of hyperscale datacentres that could be borrowed for use on-premise.
When asked about their main virtualisation and container challenges, 28% said they wanted to address virtualisation performance penalties, a quarter saw managing multi-cloud as their biggest challenge, and just over one-fifth (22%) saw security and automation as their main virtualisation challenge.
The Futuriom study also recognised that network optimisation technologies are a key way to improve datacentre performance. For the datacentre managers surveyed in the study, the potential benefits in upgrading the network include faster application performance (64%), stronger security (59%), greater flexibility (57%) and application reliability (57%).
The vast majority (84%) thought network infrastructure was either “very important” or “important” to delivering artificial intelligence and machine learning.
Read more about datacentre strategies
- Devising compute strategies for AI applications can be challenging. Find out about the hardware, network and software frameworks available.
- Converged and hyper-converged infrastructure are ready-tested ways to put compute and storage into the datacentre. But which should you choose for your deployment?