Running mission-critical applications on AWS sustainably
summary
Aerospike on AWS Graviton promises concrete reductions and measurements rather than platitudes. Customers convert a 450 node in‑memory deployment to 30 nodes from 450 and cut database latency from multiple milliseconds to 0.2 ms DB latency for many transactions, reducing SLA misses by ~30x and saving hundreds of millions annually. Lab runs show 40–60% better price performance on Graviton and prior i3en tests hit millions of TPS on 20 nodes.
AI-written summary · about
chapters (15)
- 0:00Intro
- 1:55Anatomy of a mission-critical workload
- 3:09Why run critical workloads on AWS $ Agility Team can esperiment and innovate quickly and frequently
- 5:45Broadest and deepest compute platform choice CATEGORIES CAPABILITIES OPTIONS
- 6:29Aerospike on AWS is a natural fit
- 7:41AWS Sustainability
- 9:07Sustainability: AWS Graviton Processors
- 10:10Mission-critical Use Case #1 - Threat Metrix Global Real-time Digital Identity
- 12:24Production latency comparison - Aerospike vs. Other DB
- 13:20Mission-critical Use Case #2 - Fraud Prevention
- 14:02In-memory vs. hybrid memory - SLA in deployment
- 14:53Hybrid Memory Architecture for PB Scale
- 16:07In-memory vs HMA – reducing resource footprint
- 17:12Aerospike on Graviton
- 18:02Key Takeaways
related talks
49mBuilding real-time applications to utilize AWS Graviton
07mWorking Backward From the Customer with AWS
15mAWS On Air Voice of the Customer ft. Aerospike, Inc.
22mGraviton
01h 00mfor virtually every application
25mAWS Graviton: Discover the Power of Distributed SQL + Graviton
25mAdopt AWS Graviton2 instances for best price-performance in EC2
19mFireside Chat