Sug Talk: Ellexus’ Rosemary Francis On System Telemetry And I/O Profiling For Cloud Migration And Containers

By Staff

Apr 1, 2019 | Blog

TL;DR: There is both a business and technical cost to bad I/O; fortunately, you don’t need to settle for it. Featuring Ellexus’ Rosemary Francis, this SUG video presents a constructive solution.

SUG Series Introduction

The inaugural meeting of the Singularity User Group (SUG) was held March 12-13, 2019, at the San Diego Supercomputer Center (SDSC). The event attracted diverse representation from the international advanced computing community as conveyed through the post-event press release issued jointly by SDSC and Sylabs.

Over the course of the two-day event, over 20 talks were presented by members of the Singularity user, developer, and provider community. Because SUG generated a significant amount of interest, even from those who were unable to attend, we are sharing online each of the talks presented.

SUG Talk Introduction

To quote Ellexus CEO Rosemary Francis:

Many organisations waste a huge amount of time doing bad I/O, which becomes extremely apparent in a cloud environment where you pay per time. Similarly, organisations waste money by paying for more or faster storage than they need. The answer? Profiling your applications so you know what they are doing, how they are performing and all their dependencies.

Shared recently via her blog post, Rosemary has clearly enunciated concerns carrying potentially serious business and/or technical implications – concerns applicable to cloud-related projects in general … concerns potentially complicated by the uptake of containerization. Amongst the first five contributed talks granted early acceptance for presentation at SUG, Rosemary’s was a particularly welcome contribution as Ellexus squarely addresses challenges originating with data:

Many organisations are looking to move their compute architectures to the cloud and into new technologies such as containerisation. However, dependence of legacy systems and large amounts of shared storage can make it hard to migrate to more flexible compute environments. Without a clear list of dependencies, for example, it can take a lot of time to containerise applications and to size storage appropriately for the application in the new environment.

This presentation will use industry case studies to show attendees how they can get a clear picture of their shared storage requirements in order to develop an informed go-to-cloud and containerisation strategy with I/O profiling tools from Ellexus and Singularity containers. By using system telemetry, attendees will gain an understanding of how to manage hybrid cloud architectures. We will describe techniques and best-practices for use by small teams, for corporation-wide infrastructure strategy or those just starting to look at Singularity.

As alluded to in her abstract, Rosemary presented an Ellexus case study involving a genomics pipeline. Evidently:

Through I/O profiling, the [Wellcome Sanger Institute] reduced their cloud spend by 10% and cut run time from 32 hours to 18 hours.

The ability to impact both business and technical bottom lines amplifies the value of profiling I/O on a per use case basis.

Rosemary’s talk from SUG can be found below and here. Enjoy!

Related Posts

SUG Talk: Adam Simpson on NVIDIA’s HPC Container Efforts

TL;DR: Singularity is used routinely with containers provided via the NVIDIA GPU Cloud (NGC). In this SUG talk, best practices and lessons learned are shared for the benefit of all users of containers that target execution on one or more GPUs. SUG Series Introduction...

read more

Pin It on Pinterest