Aerospike's Journey Through Innovations, Sustainability, and Roadmap
summary
Claims a patented hybrid memory architecture that stores indexes in DRAM data on SSDs, yielding near-memory latency while cutting infrastructure needs by up to 80% fewer nodes. Adds document, graph and vector models — graph supports trillions of vertices and multi-hop queries at 100,000+ QPS, vector P99 ~20 ms — with disaggregated compute/storage and DBaaS on AWS and GCP. Targets five‑nine (59) availability and TCO plus CO2 reductions as concrete outcomes.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 0:45Founding principles
- 3:16Real-time data platform
- 8:58Hybrid memory architecture
- 13:17Four customer promises
- 16:31Key use cases
- 17:57AI and real-time data
- 20:03Cloud and AI priorities
- 22:13Graph capabilities and PayPal
- 24:22Vector search integration
- 26:48Summary
related talks
31mAerospike Today and Tomorrow: Our Roadmap
32mFuture-Proofing your Architecture - AI at Any Scale with HPE & Aerospike
15mAWS On Air Voice of the Customer ft. Aerospike, Inc.
11mThe Power of Multi-Model: KV, Document, Graph
18mProduct Review and Roadmap
16mAerospike + Intel® persistent memory
31mAerospike's Innovation Agenda
30mAerospike Graph: Massively Scalable, High Performance