Ross ROSS = Recommend OSS · open-source software intelligence for agents

pravega/pravega

Pravega - Streaming as a new software defined storage primitive observed · 2026-08-28

github.com/pravega/pravega · homepage · Java · Apache-2.0 (permissive) observed · 2026-08-28

Health v2 · maintenance only

27/100

  • Activity 9
  • Release rhythm 8
  • Longevity 100
How is this computed?

round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-03. Adoption (stars, forks) is never an input.

  • gap_med: n/a
  • age_days: 3705
  • days_rel: n/a
  • days_push: 549
  • n_releases_24m: 0

Full methodology

Adoption not part of the score

1997 stars · 399 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Pravega is an open-source distributed storage service that implements Streams as a first-class primitive: durable, elastic, append-only, unbounded byte streams with strict ordering, consistency, and exactly-once semantics. It tiers data from a low-latency durable log (Tier 1, e.g., BookKeeper) to scalable long-term storage (Tier 2, e.g., HDFS, S3), and offers client APIs for event streaming, batch reads, byte streams, and key-value tables.

Use cases

  • ingest IoT sensor data into a durable event stream
  • replace Kafka with auto-scaling streams and long-term retention
  • build a Flink stream processing pipeline with exactly-once guarantees
  • store and replay video feeds for computer vision analytics
  • implement pub/sub messaging with strong ordering and consistency
  • store application state and key-value tables alongside event streams
  • unify batch and real-time access to the same data without duplication

When to choose

  • you need durable, ordered event streams with exactly-once semantics
  • your ingestion rate fluctuates and you want automatic stream scaling
  • you need both real-time streaming and batch access to historical data in one system
  • you want transparent tiering from low-latency storage to cheap cloud object storage
  • you operate at high partition counts where Kafka-style systems degrade

When to avoid

  • you need a simple, lightweight message queue with a large ecosystem of connectors
  • your team cannot operate a JVM-based cluster with ZooKeeper and BookKeeper dependencies
  • you only need short-lived pub/sub without long-term retention
  • you require non-JVM first-class client support for production workloads

Facets

service · maturity active

streaming message-queue database caching big-data microservices iot analytics jvm self-hosted cloud stream-storage event-streaming tiered-storage exactly-once append-only-log pub-sub flink-integration storage data-engineering real-time linux docker kubernetes

10 sources

Member repositories

RepositoryRoleHealth v2
pravega/pravegamain27

For agents

markdown · JSON · MCP: product_card(name="pravega/pravega")

Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem