Accepting submissions
Not accepting submissions
Arpit Bhayani Databases Were Not Designed For ThisDescription Databases were not designed for agents. They were built around a set of implicit assumptions: callers issue predictable queries, connections are short-lived, bad queries fail loudly, and schemas are a contract with engineers. Agentic systems break every one of these assumptions. Agents reason their way to queries, hold connections while an LLM thinks, retry operations unpredictably, a… more
Session type - select the format for your session: 30-minute talk – technical deep dive
|
|
V
Varuni Rethinking Data Systems for the Age of LLMsIntroduction Over the past few years, the center of gravity in data systems has begun to shift. While traditional database workloads were dominated by deterministic transactions and analytical queries, both industry and academic evidence now point to a rapid rise in AI-driven, token-based workload. Analysts estimate that 80% of enterprise data is unstructured, yet historically underutilized. Toda… more
Session type - select the format for your session: Birds of Feather (BOF) proposals – discussion on focussed topics
|
|
sm
sarthak makhija Fast on Paper, Slow in Reality: What We Got Wrong About PerformanceDescription In distributed systems engineering, a design that is “correct on paper” is only the beginning; the real challenge is making it “fast in reality.” This session offers a transparent post-mortem of the architectural assumptions we made while building a distributed key-value store from scratch in Go, and why several of those assumptions collapsed under production-grade pressure. We’ll mov… more
Session type - select the format for your session: 30-minute talk – technical deep dive
|
|
MB
Mihai Budiu Presenter Incremental ComputationIncremental computations repeatedly evaluate a function on some input values that are “changing”. The goal of an efficient implementation is to “reuse” previously computed results: when presented with a new change to the input, an incremental computation should only perform work proportional to the size of the changes of the input, rather than to the size of the entire dataset. more
Session type - select the format for your session: 30-minute talk – technical deep dive
|
|
SG
Shomik Ghose Rolling Your Own Database (Safely!): Property-based Testing at ScaleDescription There are real advantages to building a specialized database: better performance, less impedance mismatch, and lower operational cost. But the conventional wisdom against rolling your own exists for a reason. ACID is hard, and general-purpose systems are reliable precisely because they’ve been battle-hardened over decades. That calculus has recently changed - cheap object storage and … more
Session type - select the format for your session: 30-minute talk – technical deep dive
|
|
KR
Karthik Ramachandra Presenter Database systems: a decade of disruption and innovationThe past decade has been highly eventful for Database systems, to say the least. more
Session type - select the format for your session: 30-minute talk – technical deep dive
|
|
VM
Varun Mishra From Timeout to Sub-Second: Solving Scale-Dependent Deadlocks in Distributed SystemsAbstract In highly coordinated distributed systems like Apache HBase, operations often rely on global barriers, synchronized procedures that require every node to reach a consensus point before moving forward. At extreme scales, these barriers become highly sensitive to thread contention and coordination overhead. This talk details a real-world production incident at Flipkart where a critical dis… more
Session type - select the format for your session: 15-minute talk – focused engineering experience
|
|
SS
Sohham Seal Migrating Panacea.AI's 5TB/day Log Platform from Manticore to Clickhouse...and the lessons learntDescription We sized the storage layer for Nutanix’s Panacea.AI platform — 5 TB and 5 billion log lines a day — three different ways and got three answers an order of magnitude apart. Same workload, same retention, same ingest rate; the engines disagreed on storage by 37× and on CPU by 3×. more
Session type - select the format for your session: 15-minute talk – focused engineering experience
|
|
VJ
Vasudev Jamwal What Breaks When Aerospike Hits 6 Million QPSCategory: War Stories & Lessons Learned Abstract When your database is in the critical path of every ad auction, failure isn’t abstract. A misconfigured cluster costs you money in real time. A CPU spike at 1AM means your bidder is throttling while your competitors are not. more
Session type - select the format for your session: 30-minute talk – technical deep dive
|
Get your hybrid access ticket
Hosted by
We care about site reliability, cloud costs, security and data privacy