Learn how GMI achieves high performance for its databases
summary
Genomics Medicine Ireland wires sequencers to NetApp Cloud Volumes over a 10 Gbps Direct Connect, mounts dual‑protocol volumes so Windows sequencers write directly, and streams roughly 0.5 PB/month into AWS. Processing uses auto‑scaling fleets of ~100–400 EC2 instances (mostly Spot), with Lambda jobs snapshotting volumes to S3 for backup. Provisioning a new NetApp volume and on‑prem networking takes about 24 minutes to provision, and the stack delivers up to 90% cost savings versus rolling ephemeral storage while keeping data integrity and scalability.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 0:52About GMI
- 2:35What is GMI
- 10:29Retooling to AWS
- 15:44Obstacles Hurdles
- 19:06Setting up the trial
- 20:59Not done with data
- 21:45Lessons learned
- 24:45Onprem storage footprint
- 26:58Recoverability
- 30:28Sequencer
- 33:07Future plans
- 34:58Wrap up
related talks
09mThe DNA of Next-generation Data C
15mExperiences in Managing the Performance and Reliability of a Large-Scale Genomics...
29mAccelerate your cloud architectures with the right storage options
56mChoosing the right storage service for the job
16mA novel Cloud High Performance Computing infrastructure for user
12mCloud Native Application-aware Data Management in multi-cloud
29mLarge-Scale Benchmark Results of NVMe Over Fabric Technology
28mData @Scale Seattle- Brian Beach