Loading...

Xporter

XML

Word

Printable

ARTs:

Services
Benefit hypothesis:

Hide

We need to understand the practical performance of our software in order to test our architecture and understand feasibility of reaching the required performance.

Show
We need to understand the practical performance of our software in order to test our architecture and understand feasibility of reaching the required performance.
Acceptance criteria:
Hide

Report showing what limits performance of the prototype code when running on IRIS

Code and notes to reproduce the deployment and measurement which was used to learn what is presented in the report

(Stretch: improve performance)
Show
Report showing what limits performance of the prototype code when running on IRIS Code and notes to reproduce the deployment and measurement which was used to learn what is presented in the report (Stretch: improve performance)
Feature Points:
3
Initial Size:
3
WSJF:
1.667
Epic Link:
SDP Initial workflows
Agile Teams:

Team_PLANET
Due Sprint:
Sprint 5
Story Point Burn-up:
Overdue:
Outcomes:

Hide

Code in IO prototype gitlab repo

Show
Code in IO prototype gitlab repo
Resolved PI.Sprint:
19.5

The strategy is to do experiments from which something useful is learnt, then automate/document this so that can be easily repeated.

Shakedown IRIS resources for SDP workflow development use.

Create an initial description/documentation of steps required to deploy and test I/O benchmark on IRIS.

Stretch: Improvements to io-benchmark performance.

Base this on io-benchmark with Karels W-Stacking Predict implementation?

(Use of stackhpc framework for benchmarking?)

Story Point Burn-up: (57.69%)

Feature Estimate: 3.0

Deploy and test (distributed predict) I/O-benchmark on IRIS