---
title: "Performance"
description: "817 posts about Performance, summarised, each linking to the original."
---

# Performance
> 817 posts about Performance, summarised, each linking to the original.

## Articles

### [What's new in ClickHouse Managed Postgres: Customer notifications, better observability, faster backups, extensions, and more](https://yomu.fyi/post/what-s-new-in-clickhouse-managed-postgres-customer-notifications-bette.md)
- Company: [Clickhouse](https://yomu.fyi/company/clickhouse.md)
- Published: Aug 6, 2026
- Type: Announcement

ClickHouse reports a set of updates to its Managed Postgres beta, aimed at simplifying onboarding and advancing its unified Postgres-and-ClickHouse data stack. The redesigned flow supports sample-data loading or PostgreSQL migration through ClickPipes, built-in SQL exploration, replication to ClickHouse, and cross-system queries through pg\_clickhouse; services also notify users when disk usage remains above 85%, with console, email, and optional Slack channels. Observability now includes searchable PostgreSQL logs, CPU, memory, disk, IOPS, connection and per-table metrics, query insights, a Prometheus-compatible /metrics endpoint, and an OpenAPI slow-query API. Backups use more CPU and disk throughput on larger servers, making backup and restore times predictable even on 10+ TB servers, while extensions add pg\_re2 and improvements to pg\_clickhouse and pg\_stat\_ch, including regex pushdown, Arrow batching, sampling, and safer worker exits; Google Cloud Platform availability and further maintenance, configuration, and integration work are planned.


### [Introducing OfficeQA Pro V2: A New Benchmark for Enterprise Grounded-Reasoning](https://yomu.fyi/post/introducing-officeqa-pro-v2-a-new-benchmark-for-enterprise-grounded-re.md)
- Company: [Databricks](https://yomu.fyi/company/databricks.md)
- Author: Databricks AI Research Team
- Published: Aug 6, 2026
- Type: Benchmark

OfficeQA Pro V2 benchmarks whether AI agents generalize grounded reasoning to unfamiliar enterprise document collections rather than specialize in one corpus or task distribution. It contains 90 questions grounded in roughly 120,000 pages of U.S. Treasury accounting records spanning 1793–2024, including dense tables, charts, revised values, and changing reporting conventions. Questions were generated with asynth from seeded inputs, corpus evidence searches, traceable reasoning, and quality gates, then evaluated with deterministic exact match at 0.0% tolerance. Out-of-the-box frontier agents averaged 37.5% in one matched comparison, while competition systems averaged 41.1% and the winning team reached 63.3%; Databricks Genie raised matched-model accuracy by 24.0 percentage points on average, although failures remained in parsing, temporal reconciliation, and entity or category interpretation.


### [How Cursor Router chooses the right model for the task](https://yomu.fyi/post/how-cursor-router-chooses-the-right-model-for-the-task.md)
- Company: [Cursor](https://yomu.fyi/company/cursor.md)
- Author: Connor O'Keefe & Yuri Volkov
- Published: Aug 6, 2026
- Type: Explainer

Cursor Router selects models for developer turns using production behavior rather than benchmark scores, balancing user satisfaction and cost. Its routing system first uses Compass, a continuous 0-to-1 complexity predictor trained on follow-up behavior, to keep simple turns on a price-efficient model or upgrade complex ones. Then a taxonomy of domains, tasks, and modifiers compares frontier-model strengths, while an optimizer applies a one-sided 75% uplift threshold and each mode's cost budget. The dataset contains hundreds of thousands of live Cursor turns, preserving conversation signals, model-switching effects, inferred performance from subsequent user actions, and token-based costs including cache misses. Current results report Auto Intelligence at above Fable-level satisfaction with 68% lower cost, and Auto Balance outperforming Opus 4.8 at 41% lower cost while raising satisfaction by 3%.


### [ClickHouse Release 26.7](https://yomu.fyi/post/clickhouse-release-26-7.md)
- Company: [Clickhouse](https://yomu.fyi/company/clickhouse.md)
- Published: Aug 6, 2026
- Type: Announcement

ClickHouse 26.7 is a monthly release with 61 new features, 112 performance optimizations, and 329 bug fixes, including improvements to aggregation, joins, vector search, phrase search, query analysis, and URL access. Its combination of optimize\_aggregation\_in\_order and the new optimize\_aggregation\_in\_order\_limit turns matching GROUP BY, ORDER BY, and LIMIT queries into low-memory streaming pipelines that can stop reading after N groups. In a TPC-H scale-factor-100 test, the optimized query ran 313× faster, processed 339× fewer rows, and used 592× less peak memory. The release also adds build-side JOIN pruning, LLVM JIT compilation for supported regular expressions, and macOS improvements including faster allocation-heavy workloads, concurrent remote-replica reads, SSD cache dictionaries, FileLog, and STREAM support.


### [How Physical Intelligence unified its robotics data stack with Postgres managed by ClickHouse](https://yomu.fyi/post/how-physical-intelligence-unified-its-robotics-data-stack-with-postgre.md)
- Company: [Clickhouse](https://yomu.fyi/company/clickhouse.md)
- Published: Aug 6, 2026
- Type: Problem & solution

Physical Intelligence, which builds robotics foundation models, needed a data stack that could support both transactional operations and high-cardinality analytical workloads. Its previous single RDS Postgres instance could not handle those demands, so the team migrated to ClickHouse Cloud and adopted Postgres managed by ClickHouse for OLTP alongside ClickHouse for OLAP. Postgres is automatically replicated into ClickHouse through ClickPipes, while materialized views, full-text indices, and a query agent help researchers search across the company’s data. The resulting stack supports faster, easier exploration as data volume grows from the 10 billion to 100 billion row range, and also handles telemetry through ClickHouse with ClickStack and Grafana.


### [What is WAL backpressure, and why does ClickHouse Managed Postgres need it?](https://yomu.fyi/post/what-is-wal-backpressure-and-why-does-clickhouse-managed-postgres-need.md)
- Company: [Clickhouse](https://yomu.fyi/company/clickhouse.md)
- Published: Aug 5, 2026
- Type: Explainer

Postgres writes every change to the write-ahead log (WAL), and Managed Postgres uploads completed segments to object storage for point-in-time recovery; if uploads lag, retained WAL can fill the disk and cause a PANIC shutdown. ClickHouse Managed Postgres counters this with a data-plane systemd timer that checks pending segments every 15 seconds and uses cgroup v2 I/O controls to throttle client write bandwidth at 80%, 50%, or 20% of provisioned throughput as backlog thresholds rise. The drain path—including the archiver, checkpointer, and related processes—is placed in an immune cgroup, while client backends are throttled and reads remain uncapped. In a test with a 500 MB/s gp3 volume and an archive limited to 4 MB/s, the backlog reached 3,433 segments and 53 GiB before recovery. Once the limit was removed, the archiver cleared the queue, the throttle released automatically, and the disk stayed below 33% full.


### [Postgres Changes gets AND filters, new operators, and column selection](https://yomu.fyi/post/postgres-changes-gets-and-filters-new-operators-and-column-selection.md)
- Company: [Supabase](https://yomu.fyi/company/supabase.md)
- Author: filipe
- Published: Aug 5, 2026
- Type: Announcement

Postgres Changes now supports filters composed across multiple columns, additional operators, and column selection for subscriptions over Postgres tables. It can receive insert, update, and delete events over WebSockets, while comma-separated conditions or the postgresChangesFilter builder combine conditions with AND. New operators include like, ilike, is, match, imatch, isdistinct, and negated forms, while selected payloads always include the table's primary key. Column selection is opt-in, requires @supabase/supabase-js 2.109.0 or newer, and columns must be selectable by the subscribing role; Row Level Security still applies. These changes can lower billable usage because rejected events avoid messages per subscriber and trimmed payloads send fewer egress bytes, although DELETE events carry only the primary key and OR composition is unsupported.


### [Sidekick's continual learning loop](https://yomu.fyi/post/sidekick-s-continual-learning-loop.md)
- Company: [Shopify](https://yomu.fyi/company/shopify.md)
- Author: 2023-10-18
- Published: Aug 5, 2026
- Type: Explainer

Frontier models can accelerate an AI product launch, but their serving cost and latency can become impractical as usage grows, while their weights do not learn from production failures. Shopify’s proposed flywheel captures low-quality conversations, defines quality with a rubric and calibrated judges, improves the surrounding harness through autoresearch, and then converts repaired trajectories into supervised fine-tuning and GRPO training signal. A panel of frontier reasoning models critiques failures, an arbiter produces repair instructions, and unresolved cases go to expert annotators; gist compression then shortens the agent’s static prompt without measured judge-quality loss. In production, the GraphQL agent serves up to 2,000 requests per minute and reportedly achieves a 96% serving-cost reduction versus an estimated frontier-model cost, with lower latency and higher throughput. The approach is presented as a continual-learning system that turns production experience into model-weight updates.


### [How Mercado Libre rebuilt its observability platform on ClickHouse Cloud with 50x faster trace queries](https://yomu.fyi/post/how-mercado-libre-rebuilt-its-observability-platform-on-clickhouse-clo.md)
- Company: [Clickhouse](https://yomu.fyi/company/clickhouse.md)
- Published: Aug 4, 2026
- Type: Problem & solution

Mercado Libre rebuilt its O11y events observability platform to answer granular business-level questions such as why a payment failed across a large microservice environment. The previous storage-layer design performed transformation, filtering, and aggregation at query time, causing queries to run longer than five minutes and making high-cardinality analysis expensive. The team moved storage to ClickHouse Cloud on AWS while retaining its existing ingestion path, and shifted heavy work into the database with materialized views, layered retention, and a dedicated lookup table for high-cardinality filtering. O11y events uses OpenTelemetry instrumentation with business-flow attributes and samples all traffic for critical applications. The resulting system reduced trace-query time to about four seconds, a 50x speedup, achieved up to 89% compression, and scaled from 7 million to 400 million spans per minute in ClickHouse.


### [Fixed cadence to seconds: making ClickHouse Cloud autoscaling more reactive](https://yomu.fyi/post/fixed-cadence-to-seconds-making-clickhouse-cloud-autoscaling-more-reac.md)
- Company: [Clickhouse](https://yomu.fyi/company/clickhouse.md)
- Published: Aug 4, 2026
- Type: Problem & solution

ClickHouse Cloud’s vertical autoscaling recommender originally ran on a fixed timer, creating latency when workloads suddenly increased and services needed resources before the next scheduled pass. The redesign keeps the periodic sweep as a safety net while adding a reactive fast path, using controller-runtime as an event-processing layer with sources feeding a shared rate-limited work queue and reconciler. The queue provides deduplication, exponential backoff, bounded concurrency, leader election, and metrics, while ClickHouse stores near-real-time signals that a polling source reads; the excerpt notes polling sets a floor on reaction speed. The reactive path is rolling out cautiously in observe-only mode, with periodic reconciliation remaining the main driver, and the stated goal is to trigger scale-ups for critical events such as out-of-memory conditions in seconds rather than waiting for the next scheduled run.


### [Bringing serverless functions closer to the speed of wire](https://yomu.fyi/post/bringing-serverless-functions-closer-to-the-speed-of-wire.md)
- Company: [Modal](https://yomu.fyi/company/modal.md)
- Published: Aug 4, 2026
- Type: Announcement

Modal’s legacy I/O plane routed all function inputs and outputs through us-east servers, adding up to ~200ms of network latency for calls from other regions. The new geographically distributed I/O plane, available in four or more regions, reduces end-to-end function call latencies by roughly 80ms at p50. The redesign involved moving non-critical work to background async processing, minimizing shared storage interactions on the hot path, and rewriting the server layer in Go for better concurrency. The team evaluated Valkey but chose Redis 7.1 because engine CPU usage spiked aggressively on Redis 7.2 onwards during input queue load testing. Users can further optimize by selecting the nearest routing\_region, keeping input and output payloads under 2MiB to avoid blob storage uploads, and batching small inputs to reduce network hops.


### [NVIDIA Vera Storage Benchmarks: Faster Encryption, Compression, Integrity Checking, and Recovery for AI-Native Storage](https://yomu.fyi/post/nvidia-vera-storage-benchmarks-faster-encryption-compression-integrity.md)
- Company: [NVIDIA](https://yomu.fyi/company/nvidia-developer-blog.md)
- Author: Elizabeth Goodman
- Published: Aug 3, 2026
- Type: Benchmark

Agentic AI architectures require storage systems to secure, compress, and validate massive streams of persistent data and key-value cache entries across concurrent reasoning loops. These tasks run primarily on CPUs and frequently create processing bottlenecks that prevent modern networks and flash storage from operating at full capacity. NVIDIA evaluated the Vera CPU within the BlueField-4 STX Storage Processor using a purpose-built in-memory microbenchmark suite running standard libraries like OpenSSL, Zstandard, and LZ4. The results demonstrated that Vera outperformed x86 processors across core storage primitives including encryption, decompression, Reed-Solomon recovery, and CRC32C integrity checking. For a sequential multi-stage pipeline executing compression followed by encryption, Vera achieved up to 3.21x higher throughput than the comparison x86 CPU.


### [Ingest semi-structured data faster and more efficiently with Variant - Now Generally Available](https://yomu.fyi/post/ingest-semi-structured-data-faster-and-more-efficiently-with-variant-n.md)
- Company: [Databricks](https://yomu.fyi/company/databricks.md)
- Author: Jonathan Brito, Gene Pang, Harsh Motwani
- Published: Aug 3, 2026
- Type: Announcement

Databricks announces general availability of the Variant data type for ingesting semi-structured JSON, XML, and CSV while avoiding the usual choice between flexible storage and fast, schematized queries. Variant Shredding, also generally available, stores common fields as columns in Parquet files, while Predictive Optimization uses workload and query patterns to identify important fields, collect statistics, and improve file skipping. More than 5,000 teams write Variant, with users executing over 500 million Variant queries monthly across more than 160 TB of data; shredding delivers nearly four-times-faster reads than unshredded Variant and 30-times-faster reads than JSON strings. Auto Loader and Zerobus can ingest Variant into Delta or Iceberg, and future plans include Liquid Clustering by Variant fields, expanded SQL functions, and additional integrations.


### [ClickHouse launches ClickHouse Labs with Andy Pavlo as VP of Database Research](https://yomu.fyi/post/clickhouse-launches-clickhouse-labs-with-andy-pavlo-as-vp-of-database.md)
- Company: [Clickhouse](https://yomu.fyi/company/clickhouse.md)
- Published: Aug 3, 2026
- Type: Announcement

ClickHouse announced the launch of ClickHouse Labs, a new research group led by Andy Pavlo, who joins as Vice President of Database Research. The group is intended to advance foundational work in database systems and share it openly with the broader community. Its research spans database and data architectures, query processing, systems performance, and infrastructure for both ClickHouse and PostgreSQL, with applications including real-time analytics, data warehousing, observability, and AI. Researchers will build and test ideas against demanding real-world workloads, collaborate with ClickHouse engineers, and publish findings in leading research venues so others can evaluate, extend, and build on them.


### [Andy Pavlo joins ClickHouse to establish ClickHouse Labs](https://yomu.fyi/post/andy-pavlo-joins-clickhouse-to-establish-clickhouse-labs.md)
- Company: [Clickhouse](https://yomu.fyi/company/clickhouse.md)
- Published: Aug 3, 2026
- Type: Announcement

Andy Pavlo announces that he is joining ClickHouse to establish and lead ClickHouse Labs, an industry research team focused on database systems. He describes ClickHouse’s early use of C++ and vectorized query execution with SIMD in 2016, and says its development has remained relevant to his academic research. Labs will work directly with ClickHouse engineers, customers, collaborators, and industry partners rather than operating separately, while also collaborating with the PostgreSQL team on performance and reliability for its managed service. An immediate priority is validating and pushing an existing backlog of ideas and optimizations into production, followed by research into hardware, algorithms, data structures, execution strategies, and DBMS development for AI and agentic workloads. The stated aim is to combine scientific research with technology that matters to users.


### [Backstage with Lakebase, part 3](https://yomu.fyi/post/backstage-with-lakebase-part-3.md)
- Company: [Databricks](https://yomu.fyi/company/databricks.md)
- Author: Cameron Casher, Shanil Anushka Fernando, Kevin Hartman
- Published: Jul 31, 2026
- Type: Explainer

This third part describes joining Backstage’s live operational Postgres catalog with analytical billing data so teams can query resource ownership and cloud spend together. Lakebase isolates compute per workload, allowing portal traffic and FinOps queries to use the same underlying storage; benchmarked catalog queries ran at 55–65 ms end-to-end and searches at two to four ms. Because Lakehouse Federation’s Postgres connector accepts static credentials while Lakebase applications use OAuth JWTs, the proof-of-concept creates a separate SCRAM-SHA-256 Postgres role for federation. This enables a single query to join Backstage resource names with Unity Catalog’s system.billing.usage without data movement, while branch billing remains visible per branch and endpoint. The operational considerations include password rotation, read-only grants, branch TTLs, and scale-to-zero endpoints that stop billing when idle.


### [Co-Designing AI Model Attention for Fast, Interactive Long-Context Inference](https://yomu.fyi/post/co-designing-ai-model-attention-for-fast-interactive-long-context-infe.md)
- Company: [NVIDIA](https://yomu.fyi/company/nvidia-developer-blog.md)
- Author: Tanya Lenz
- Published: Jul 31, 2026
- Type: Explainer

As context lengths increase in long-context and agentic workloads, attention increasingly dominates GPU inference runtime. Dense attention performance depends on architectural parameters like group size, head dimension, and sequence length, alongside GPU parallelization strategies. Prefill remains compute-bound and scales quadratically with input sequence length, showing minimal runtime sensitivity to group size variations. Conversely, decode generation is memory-bound and scales linearly with KV cache length, where doubling group size roughly halves decode execution time. To maximize throughput and interactivity on NVIDIA hardware, model architects should push group sizes higher, select head dimensions of 128 or 256, reduce effective KV state, and keep tensor parallelism less than or equal to the number of KV heads.


### [Don’t stop early: Case-folding source code at memory speed](https://yomu.fyi/post/don-t-stop-early-case-folding-source-code-at-memory-speed.md)
- Company: [Github](https://yomu.fyi/company/github.md)
- Author: Alexander Neubeck
- Published: Jul 31, 2026
- Type: Problem & solution

GitHub's code search engine, Blackbird, must case-fold over 480TB of source code across 180 million repositories during indexing and query matching. To accelerate this operation on source code that is overwhelmingly ASCII, the engineering team replaced early-exit branching with an unconditional branch-free loop. The implementation tests uppercase ASCII ranges using wrapping arithmetic, modifies bits in place, and detects non-ASCII bytes with an accumulator register tested only after the loop completes. Eliminating data-dependent exits allowed LLVM to generate SIMD instructions and achieve throughput exceeding 45 GiB/s on an Apple M4 processor. The optimized implementation was released as the open-source Rust crate casefold.


### [NVIDIA Video Codec SDK 13.1: Zero-Copy Transcode, AV1 B-Frames, and Frame-Accurate Seek](https://yomu.fyi/post/nvidia-video-codec-sdk-13-1-zero-copy-transcode-av1-b-frames-and-frame.md)
- Company: [NVIDIA](https://yomu.fyi/company/nvidia-developer-blog.md)
- Author: Elizabeth Goodman
- Published: Jul 31, 2026
- Type: Announcement

Modern video pipelines require higher efficiency and support for complex workloads across streaming, collaboration, and artificial intelligence media tools. NVIDIA released Video Codec SDK 13.1 to expand GPU-accelerated encoding, decoding, and transcoding capabilities. The release introduces Hierarchical Reference Mode for AV1 with up to 31 B-frames, combines UHQ tuning with iterative encoding, and extracts per-macroblock decode statistics for H.264 and HEVC without CPU overhead. It also implements frame-accurate, GOP-aware seeking through the NvVideoDecoder class and adds stereoscopic MV-HEVC decode metadata. To simplify setup across different environments, the update delivers an official multi-stage Docker container pinned to Ubuntu 22.04 LTS, CUDA 12.3.2, and Vulkan SDK.


### [Run High-Performance Core Math at Scale with NVIDIA nvmath-python](https://yomu.fyi/post/run-high-performance-core-math-at-scale-with-nvidia-nvmath-python.md)
- Company: [NVIDIA](https://yomu.fyi/company/nvidia-developer-blog.md)
- Author: Michelle Horton
- Published: Jul 30, 2026
- Type: Announcement

NVIDIA released nvmath-python v1.0 to give Python users direct access to CUDA-X math libraries without requiring low-level C or C++ interfaces. The library acts as an abstraction layer over GPU engines like cuFFT and cuBLASLt, CPU backends like NVPL and Intel MKL, and distributed libraries such as cuBLASMp. It interfaces directly with NumPy, CuPy, and PyTorch tensors, structuring routines into generic interfaces for broad interoperability and specialized advanced modules for hardware-specific optimizations. Users can define custom sparse formats with Universal Sparse Tensors or invoke compiled device APIs directly inside custom numba-cuda kernels. Built-in logging captures execution space decisions and flags potentially expensive memory transfers across host and device environments.


[Newer posts](https://yomu.fyi/topic/performance/page/8.md) · [Older posts](https://yomu.fyi/topic/performance/page/10.md)