Search Results for "throughput" - Page 2
Sort By:
Showing 502 open source projects for "throughput"
View related business solutions-
Streamline Azure Security with Palo Alto Networks VM-SeriesImprove your security posture and reduce incident response time. Use the VM-Series to natively analyze Azure traffic and dynamically drive policy updates based on workload changes.
-
Earn up to 16% annual interest with Nexo.Put idle assets to work with competitive interest rates, borrow without selling, and trade with precision. All in one platform. Geographic restrictions, eligibility, and terms apply.
-
1
Motor
The async Python driver for MongoDB and Tornado or asyncio
...It provides a familiar API surface similar to the official synchronous PyMongo driver, so you can migrate or write MongoDB code in Python without having to learn a completely new interface. Because it integrates with popular async ecosystems like FastAPI, Sanic, and aiohttp, Motor is a natural fit for modern async Python stacks where throughput and responsiveness matter. It also supports change streams, grid file system (GridFS), and the full range of CRUD and aggregation operations available in MongoDB. -
2
CoreNet
CoreNet: A library for training deep neural networks
...CoreNet provides abstractions for data, tensor, and pipeline parallelism, allowing models to scale without code duplication or heavy manual configuration. Its distributed runtime manages synchronization, load balancing, and mixed-precision computation to maximize throughput while minimizing communication bottlenecks. CoreNet integrates tightly with Apple’s proprietary ML stack and hardware, serving as the foundation for research in computer vision, language models, and multimodal systems within Apple AI. The framework includes monitoring tools, fault tolerance mechanisms, and efficient checkpointing for massive training runs. -
3
ConcourseDB
Distributed database warehouse for transactions, search and analytics
ConcourseDB is a distributed, self-tuning database designed for real-time applications, offering strong consistency and ACID compliance without requiring complex configurations. It provides dynamic schema support and automatic indexing. -
4
FramePack
Lets make video diffusion practical
FramePack explores compact representations for sequences of image frames, targeting tasks where many near-duplicate frames carry redundant information. The idea is to “pack” frames by detecting shared structure and storing differences efficiently, which can accelerate training or inference on video-like data. By reducing I/O and memory bandwidth, datasets become lighter to load while models still see the essential temporal variation. The repository demonstrates both packing and unpacking... -
$300 in Free Credit Towards Top Cloud ServicesStart your project in minutes. After credits run out, 20+ products include free monthly usage. Only pay when you're ready to scale.
-
5
BPB-Worker-Panel
A GUI Panel providing Worker subscriptions for VLESS
BPB-Worker-Panel is a web interface for monitoring and managing distributed workers and task queues. It provides real-time statistics on job statuses, worker performance, queue lengths, and logs. Designed for systems using task handlers like Celery, Sidekiq, or custom job runners, the panel helps administrators visualize worker load, inspect tasks, and manually retry or delete them as needed. -
6
RTP-LLM
Alibaba's high-performance LLM inference engine for diverse apps
RTP-LLM is an open-source large language model inference acceleration engine developed by Alibaba to provide high-performance serving infrastructure for modern LLM deployments. The system focuses on improving throughput, latency, and resource utilization when running large models in production environments. It achieves this by implementing optimized GPU kernels, batching strategies, and memory management techniques tailored for transformer inference workloads. The framework is designed for large-scale AI services and is already used internally across several Alibaba platforms such as Taobao, Amap, and other business systems that rely on conversational or search-related AI services. ... -
7
Conscrypt
Conscrypt is a Java Security Provider
...Because it builds on BoringSSL’s audited primitives and constant-time implementations, it reduces the risk of subtle side-channel issues while improving handshake and bulk crypto throughput. Conscrypt also exposes platform accelerations (like AES-GCM and ChaCha20-Poly1305) and supports features important to HTTP/2 and HTTP/3 stacks. Its goal is pragmatic: deliver a fast, well-maintained TLS provider that behaves like modern browsers and mobile stacks. -
8
LMCache
Supercharge Your LLM with the Fastest KV Cache Layer
...Instead of rebuilding KV states for repeated or shared text segments, LMCache persists and retrieves them from multiple tiers—GPU memory, CPU DRAM, and local disk—then injects them into subsequent requests to reduce TTFT and increase throughput. Its design supports reuse beyond strict prefix matching and enables sharing across serving instances, improving efficiency under real multi-tenant traffic. The broader project includes examples, tests, a server component, and public posts describing cross-engine sharing and inter-GPU KV transfers. These capabilities aim to lower latency, cut GPU cycles, and stabilize performance for production workloads with overlapping prompts or retrieval-augmented contexts. ... -
9
DeepEP
DeepEP: an efficient expert-parallel communication library
DeepEP is a communication library designed specifically to support Mixture-of-Experts (MoE) and expert parallelism (EP) deployments. Its core role is to implement high-throughput, low-latency all-to-all GPU communication kernels, which handle the dispatching of tokens to different experts (or shards) and then combining expert outputs back into the main data flow. Because MoE architectures require routing inputs to different experts, communication overhead can become a bottleneck — DeepEP addresses that by providing optimized GPU kernels and efficient dispatch/combining logic. ... -
Full-stack observability with actually useful AI | Grafana CloudBuilt on open standards like Prometheus and OpenTelemetry, Grafana Cloud includes Kubernetes Monitoring, Application Observability, Incident Response, plus the AI-powered Grafana Assistant. Get started with our generous free tier today.
-
10
Infinity
Low-latency REST API for serving text-embeddings
Infinity is a high-throughput, low-latency REST API for serving vector embeddings, supporting all sentence-transformer models and frameworks. Infinity is developed under MIT License. Infinity powers inference behind Gradient.ai and other Embedding API providers. -
11
Reliable PostgreSQL Backup & Restore
Reliable PostgreSQL backup and restore
...Compression is usually the bottleneck during backup operations but, even with now-ubiquitous multi-core servers, most database backup solutions are still single-process. pgBackRest solves the compression bottleneck with parallel processing. Utilizing multiple cores for compression makes it possible to achieve 1TB/hr raw throughput even on a 1Gb/s link. More cores and a larger pipe lead to even higher throughput. A custom protocol allows pgBackRest to backup, restore, and archive locally or remotely via TLS/SSH with minimal configuration. An interface to query PostgreSQL is also provided via the protocol layer so that remote access to PostgreSQL is never required, which enhances security. -
12
Aeron
Efficient UDP unicast, UDP multicast, and IPC message transport
...Aeron Cluster provides support for fault-tolerant services as replicated state machines based on the Raft consensus algorithm. Performance is the key focus. A design goal for Aeron is to be the highest throughput with the lowest and most predictable latency of any messaging system. Aeron integrates with Simple Binary Encoding (SBE) for the best possible message encoding and decoding performance. Many of the data structures used in the creation of Aeron have been factored out to the Agrona project. -
13
Olric
Distributed, in-memory key/value store and cache
A lightweight, distributed in-memory data store designed for key-value caching and ephemeral storage. -
14
PikiwiDB
Pika is a Redis-Compatible database
A high-performance, distributed key-value database inspired by Redis but designed for large-scale cloud-native environments. -
15
D++
C++ Discord API Bot Library - D++ is Lightweight and scalable
D++ is a lightweight and simple library for Discord written in modern C++. It is designed to cover as much of the API specification as possible and to have an incredibly small memory footprint, even when caching large amounts of data. It is created by the developer of TriviaBot and contributed to by a dedicated team of developers. -
16
FlashMLA
FlashMLA: Efficient Multi-head Latent Attention Kernels
FlashMLA is a high-performance decoding kernel library designed especially for Multi-Head Latent Attention (MLA) workloads, targeting NVIDIA Hopper GPU architectures. It provides optimized kernels for MLA decoding, including support for variable-length sequences, helping reduce latency and increase throughput in model inference systems using that attention style. The library supports both BF16 and FP16 data types, and includes a paged KV cache implementation with a block size of 64 to efficiently manage memory during decoding. On very compute-bound settings, it can reach up to ~660 TFLOPS on H800 SXM5 hardware, while in memory-bound configurations it can push memory throughput to ~3000 GB/s. ... -
17
SoftEther VPN
Cross-platform multi-protocol VPN software
An open-source free cross-platform multi-protocol VPN program, as an academic project from University of Tsukuba, under the Apache License 2.0. The API Suite allows you to easily develop your original SoftEther VPN Server management application to control the VPN Server (e.g. creating users, adding Virtual Hubs, disconnecting a specified VPN sessions) from JavaScript, TypeScript, C# or other languages. SoftEther VPN ("SoftEther" means "Software Ethernet") is one of the world's most powerful... -
18
Zeebe
Distributed Workflow Engine for Microservices Orchestration
Automate processes at scale with unprecedented performance and resilience. Zeebe is the workflow and decision engine that powers Camunda Platform 8. Zeebe’s cloud-native design provides the performance, resilience, and security enterprises need to future-proof their process orchestration efforts. Zeebe distributes data across all brokers in a cluster with storage directly on the server filesystem. If one broker goes down, another can replace it with no data loss. This pre-configured... -
19
Node RDKafka
Node.js bindings for librdkafka
A high-performance Node.js client for Apache Kafka, built on top of librdkafka, providing bindings for efficient Kafka message processing. -
20
WireGuard Linux
Linux kernel implementation of the WireGuard VPN protocol
...The repository serves as a mirror of the official development tree and contains the kernel-side code that enables native WireGuard support on Linux systems. By operating inside the kernel rather than user space, the project achieves very low latency and high throughput, making it suitable for servers, embedded systems, and enterprise networking environments. WireGuard itself is designed to be a modern, streamlined VPN that uses state-of-the-art cryptography while remaining simpler and leaner than legacy solutions such as IPsec or OpenVPN. The Linux implementation integrates tightly with standard networking tools and interfaces, allowing administrators to manage tunnels using familiar workflows. -
21
RabbitMQ Server
Core server and tier 1 (built-in) plugins
...Supports multiple messaging protocols, message queuing, delivery acknowledgement, flexible routing to queues, multiple exchange type. Deploy as clusters for high availability and throughput; federate across multiple availability zones and regions. HTTP-API, command line tool, and UI for managing and monitoring RabbitMQ. -
22
DeepSpeed MII
MII makes low-latency and high-throughput inference possible
MII makes low-latency and high-throughput inference possible, powered by DeepSpeed. The Deep Learning (DL) open-source community has seen tremendous growth in the last few months. Incredibly powerful text generation models such as the Bloom 176B, or image generation model such as Stable Diffusion are now available to anyone with access to a handful or even a single GPU through platforms such as Hugging Face. -
23
Grafana Alloy
OpenTelemetry Collector distribution with programmable pipelines
Grafana Alloy is an open source OpenTelemetry Collector distribution with built-in Prometheus pipelines and support for metrics, logs, traces, and profiles. Grafana Alloy is Grafana Labs’ distribution of the OpenTelemetry Collector. It is an OTLP-compatible collector with built-in Prometheus optimizations that also support signals across metrics, logs, traces, and profiles. Alloy was started at Grafana Labs and announced at GrafanaCON in 2024. The mission of the project is to create the best... -
24
r2dbc-postgresql
Postgresql R2DBC Driver
r2dbc-postgresql is a reactive PostgreSQL driver for the JVM based on the R2DBC (Reactive Relational Database Connectivity) specification. It allows non-blocking, asynchronous access to PostgreSQL databases, making it ideal for reactive Java or Kotlin applications built with frameworks like Spring WebFlux. -
25
async-std
Async version of the Rust standard library
async-std makes async programming foundations easy and approachable. Built-in Rust for speed and safety.
