Sorry, you need to enable JavaScript to visit this website.
Submitted by Anonymous (not verified) on

Computational storage in general can bring unique benefits in increasing the efficiency of CPU utilization in a data processing system. In this presentation we discuss the benefits of leveraging computational storage for offloading compute intensive tasks of object storage applications in a disaggregated storage environment. We demonstrate the ability of the solution to complement the CPU by taking away tasks that benefit from in-situ processing within the storage, thereby improving the overall system performance while lowering the TCO. This model has additional benefits of reducing data movement, resulting in better network utilization and decreased latency for supporting data analytics. Disaggregated storage is particularly attractive when using computational storage since scaling storage naturally yields to scaling of tasks that can be accelerated. We discuss results from our experimentation accelerating the S3 Select functionality and other use cases, using our High Performance Object Storage (HPOS) platform.

Bonus Content
Off
Presentation Type
Presentation
Learning Objectives
  • Understanding the performance improvements that can be realized by using computational storage for accelerating compute intensive tasks.
  • Understanding the advantages of using disaggregated storage model with computational storage devices in improving the scalability of solutions.
  • Understanding the TCO and power benefits from using computational storage devices.
Display Order
70
Start Date/Time
End Date/Time
YouTube Video ID
GmkWN4pHoEY
Zoom Meeting Completed
Off
Main Speaker / Moderator
Room Location
Salon IV
Webform Submission ID
547