P99 CONF is the industry event for high performance, low latency, high availability engineering.
Joined May 2021
- Tweets 2,391
- Following 182
- Followers 10,337
- Likes 1,752
1,463 Photos and videos
LLMs are all the buzz, but they're also driving massive GPU demand and cost. At #P99CONF, @PliopsLtd's Eshcar Hillel shared a generic architecture for offloading KV-cache tensors to a disaggregated shared store. Check it out on-demand here: ow.ly/lISl50Xkpw3
#ScyllaDB
2
219
Faced with the need for resource mutation beyond Gatekeeper’s capabilities, Adevinta explored Kyverno’s out-of-the-box support for both validation and mutation. Tanat Lokejaroenlarb shared their best practices at #P99CONF. Watch here > ow.ly/bviY50XGpi7
#ScyllaDB
202
How can you make an open-source time-series database a much faster analytical one as well? @AndreyPechkurov explained how during a #P99CONF session on the journey of making #QuestDB. ow.ly/Z4wv50Uc2S5
#ScyllaDB #database
1
227
Rivian vehicles stream 5,500 signals every 5 seconds, but only about 80 are relevant for push notifications. Marcus Kim and Saahil Khurana shared how Mega Filter reduced load across Kafka, Flink, and EKS in this highly technical #P99CONF talk. ow.ly/l87L50Y9qpG
#ScyllaDB
231
Tiket manages billions of hotel inventory rows, and that pushed PostgreSQL to its limits. After migrating to #ScyllaDB, they saw 200x higher write throughput & 60% lower costs. Abdurohman Abdurohman explained how they did it @P99CONF > ow.ly/MXJW50XHxLc
#ScyllaDB #P99CONF
1
3
610
One of our more technical #P99CONF sessions dove into how ScyllaDB prevents overload by throttling clients to the optimal request rate — keeping memory stable without wasting capacity. Watch Nadav Har’El's session for free here > ow.ly/liPG50Y8Ktv
#ScyllaDB
3
273
The core and the basis of #ScyllaDB’s outstanding performance is the Seastar framework, and the core and the basis of seastar is its event loop. At #P99CONF, Pavel Emelyanov explained what the loop does, its limitations, and how to understand its behavior. ow.ly/64hG50XuOsH
1
5
313
📖 Your Weekend Read: Queues are a popular solution to handling sudden traffic spikes without crashing. But as @P99CONF speaker @p_mbanugo explains, they don't fix overload. Learn why and what to do instead. ow.ly/pZOe50Z83b5
#ScyllaDB #P99CONF
2
10
447
DragonflyDB told B trees to do sorted sets and they nailed it. Joe Zhou shares lessons learned on trade-offs, latency wins, and the quirks of adapting traditional database-born data structures in this #P99CONF on-demand session. ow.ly/hcWY50Xu9vH
#ScyllaDB
1
5
293
During a traffic spike, the team's reaction is almost always, we need a bigger buffer! #P99CONF 2026 speaker @p_mbanugo explains why queues only absorb variance and aren't a long-term solution to fixing the problem. ow.ly/CZI450Z7BPv
#ScyllaDB
1
2
225
If you're moving from C to #Rustlang, @ClickHouseDB, an open source analytics-oriented database system composed of 1.5 million lines of code, has indispensable advice. Check out what Alexey Milovidov shared at #P99CONF via @thenewstack. ow.ly/ce0q50XqAHX
#ScyllaDB
3
225
Maximizing the KV Cache hit rate is critical to minimize time-to-first-token latency. In John Thomson's #P99CONF talk, the techniques NVIDIA’s TensorRT-LLM uses to maximize KV Cache hit rates for structured LLM workloads was shared. Watch here > ow.ly/SIue50Xm5Ps
#ScyllaDB
1
5
250
📣 Calling all developers and engineers with an interest in high-performance, low-latency applications. Registration for our free and virtual #P99CONF is open and there's so much we have in store for this year's conference. Learn more here > ow.ly/Sbrw50Z6bxP
#ScyllaDB
1
2
292
pgBadger's "raw output" feature allows you to build new tools to root out the causes of performance degradation. In a highly technical #P99CONF session, @HettieDombr explained how to uplevel pgBadger so you can unlock deeper insights. ow.ly/wNM950XJC4P
2
312
How can lock-free programming and precise ultra low latency pipelining between CPU cores improve performance? Find out in Sonia Kolasinska's #P99CONF session. ow.ly/qiqJ50UtAJb
#ScyllaDB
2
11
531
Want to join the lineup of highly-technical #P99CONF speakers? May 29 is the LAST day to submit your proposal to speak at our 2026 virtual event. Thousands attend our vendor-neutral conference each year, and 2026 will be the biggest yet > ow.ly/jzKO50Z59zy
#ScyllaDB
1
209
At our last #P99CONF, @tanelpoder launched the prod/GA ready version of xCapture v3 xtop and demonstrated how it helps with real life troubleshooting scenarios. Watch the entire session for free: ow.ly/QfGR50XpsSN
#ScyllaDB
1
6
606
In John Thomson's highly technical #P99CONF talk, the techniques NVIDIA’s TensorRT-LLM uses to maximize KV Cache hit rates for structured LLM workloads were revealed. Get proven strategies for latency-critical LLM applications in this session > ow.ly/cQGc50XHy9w
#ScyllaDB
3
271
💪 Monday motivation: @P99CONF is designed to highlight the engineering challenges and solutions required for low-latency, high-performance distributed computing applications. If you have experience, submit your topic before May 29 > ow.ly/7azw50Z3Lwp
#ScyllaDB #P99CONF
3
4
805