GDB-Engines

The go-to resource for graph technology details

About
Back to all databases

ZipG

A research prototype graph store that executes queries directly on a compressed graph representation for memory efficiency. Lacks traditional database features such as transactions and replication.

GitHub 13 stars
Overall rank
#80
First released
2017
License
Not listed
Built with
C++

Details

Type
Other
Kind
Embedded
Category
Emerging
Status
Inactive — Academic research prototype from AMPLab Berkeley. No active development.
License
-
Language
C++
Query Languages
Custom API
Protocols
Thrift

Feature scores

Community & Business

Active development
0%
No updates in 6+ months
Commercial support
0%
No commercial support
Live community
0%
No community activity data
Open Source
✓
Open source license
Pricing
0%
No pricing information
Trendiness
0%
Declining trend

Deployment

Containerization
0%
No container image
Work as dedicated instance
✓
Can run as standalone server
Work as embedded
0%
Standalone only
Testing in-memory version
0%
No in-memory option

Platform

Operating on Linux
✓
Linux officially supported
Operating on Windows
75%
Open source (Windows compatible)
SaaS offering
0%
Self-hosted only

Operations

Automatic updates
0%
Manual updates only
Client side caching
0%
No client-side caching
Data versioning support
0%
No versioning
Live backups
0%
No live backups

Distribution

Cluster Re-balancing
0%
No auto-rebalancing
Data Distribution
✓
Data sharding supported
High-Availability
0%
No HA support
Query Distribution
✓
Distributed query execution
Replication support
0%
No replication

Developer Experience

Data types defined
50%
Basic data types only
Logging/Auditing
0%
No logging/auditing
Object-Graph Mapper
✓
OGM/ORM available
Reactive programming
0%
No reactive programming
Documentation up-to-date
✓
Docs current with latest version
Binary protocol
0%
Text-based protocols only
CLI
0%
No CLI
GUI
0%
No GUI

Data Model

Multi-database
0%
Graph-only
Graph-native data
✓
Native graph storage
REST API
0%
No REST API
Query Language
0%
No query language

Transactions

Granular locking
0%
No granular locking
Multiple isolation levels
0%
Read uncommitted only
Read committed transaction
0%
Not available
Transaction support
0%
No transactions

Schema & Security

Constraints
0%
No constraints
Schema support
0%
Schema-less
Secondary indexes
0%
Primary index only
Server side procedures
0%
No stored procedures
Triggers
0%
No triggers
Authentication
0%
No authentication
Authorization
0%
No authorization
Data encryption
0%
No encryption
Sources 5 references

Protocols Thrift

Aggregator exposes a GraphQueryAggregatorService via Apache Thrift; the reference Python client connects over TSocket/TBinaryProtocol to port 11001 by default; research prototype, undocumented beyond source.

  • Entry point to prepare a cluster. Performs the following in parallel:
    rpc/thrift/succinct_graph.thrift · reviewed 2026-08-02
  • from thrift import Thrift from thrift.transport import TSocket from thrift.transport import TTransport from thrift.protocol import TBinaryProtocol
    client/py/preamble.py · reviewed 2026-08-02

Released 2017

Matches current value; first public documentation is the SIGMOD 2017 paper. The GitHub mirror's creation date (2020) postdates this and is not reliable evidence of first availability, and an earlier VLDB 2015 submission (per the repo's default branch name) was reviewer-only, not public.

Language C++

Corrects current value of Java: the core graph store (SuccinctGraph, GraphSuffixStore, etc.) is implemented in C++; Java/Scala are only the client library and LinkBench benchmark harness.

github_url https://github.com/amplab/zipg

pdclab/zipg returns 404, as does anuragkh/zipg (the account its own README instructs users to clone from). amplab/zipg, under the original AMPLab organization that produced the research, is live and returns 200.