Your business could anticipate the activity of an Entity (think - customer,smart phone component, employee, etc) in your organization

Your team could spend more time exploring entity relationships rather being buried in heterogeneous data problems?

EinsteinDB's Causal Set: Causets are a closed-loop, self-learning system.

More About Us
MapShape

Our Brand Partners

Partner
Partner
Partner
Partner
Partner
Partner
View All

EinsteinDB can be run on multiple servers in widely separated locations.

All sync operations happen in a context of an InProgress - an internal EinsteinDB Causet transaction representation. If sync succeeds, all necessary operations are comitted to the underlying database in a single SQLite transaction

FunIcon

5000 Threads with 4.99 95% latency

Kubernetes allows FIDel replica Pods to run in Host network mode. This way of deployment is suitable when a MilevaDB instance occupies the whole machine without causing any Pod conflict. The Point Select test is conducted in both modes respectively

FunIcon

456,735.42 QPS

EinsteinDB is 3.29× and 2.37× faster than RocksDB-L0-NVM and NoveLSM on the load workload comprising of mostly random writes.

FunIcon

99%

200 avg tail latency.

FunIcon

2.56%

Write Amplification writing 80GB datasets.

Cloud Native Database systems

EinsteinDB processes up to 491,000 sale transactions per second, which translates to more than 70 million transactions per second.

Service Icon

Cross Shard Transactions

When we simply deploy a MySQL or PostgreSQL on a cloud instance store with a local SSD and a high I/O VM, the resulting database instance has limited capacity

Learn More
Service Shape
Service Icon

Performance-Critical

MilevaDB is a real-time OLAP database system designed for high-concurrency, low-latency, and real-time analytical queries at PB scale. It has been running on from as little as 3 nodes to up to 2000+ physical machines and is provided as a database service on AWS, Google Cloud offerings, and Azure. I

Learn More
Service Shape
Service Icon

FIDel: Self-Driving Interlocking Federated Directorate

Till now, EinsteinDB has been deployed on FIDel and applied to more than 10,000 database instances. We have successfully reduced the memory consumption by more than 17% (≥ 27TB)

Learn More
Service Shape
Service Icon

OLTP processing

the OLTP process offered by EinsteinDB, VioletaBFT, MilevaDB, FIDel, BerolinaSQL, and Noether create its working set of updated pages on demand. This is somewhat analogous to swapping pages into a buffer pool – however, the copy on demand of updated pages is three to four orders of magnitude faster as it takes only 2 µs to copy a main memory page instead of 10ms

Learn More
Service Shape
Service Icon

Multiple OLAP Sessions

s. The EinsteinDB architecture allows for arbitrarily current snapshots. This can simply be achieved by periodically (or on demand) fork()-ing a new snapshot and thus starting a new OLAP query session process.

Learn More
Service Shape
Service Icon

Multi-Threaded OLTP Processing

EinsteinDB is compatible with common rDMA interfaces (e.g., Myrinet or Infiniband) in order to unburden the server’s CPU from the data transmission task.

Learn More
Service Shape

What Our Clients are Saying?

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna.

testimonials
testimonials

EinsteinDB is a distributed tuple store. On top of it companies are building databases that speak different dialects.

Lenin will synchronize a local EinsteinDB append-only version of database against a remote replica server, modifying local state if necessary, and uploading changes to the server if necessary. Schemaless aggregates are CA for HA (adding vocabulary).

an agreement between local(spacelike) and remote states(future: lightlike or past:timelike).
last-synced local transaction
local fast-forward
remote fast-forward
Merge Append
atomic sync reports
Enterprise Knowledge Graph
one-to-many and many-to-many relations are modeled directly, instead using intermediate tables
Vector Shape
Scientist
Scientist
Scientist
Map Shape
Vector Shape
Star IconOur Team

Meet Our Data Scientist Preparing For Your Business Success

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna.

Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna.

View Our Team
Vector Shape

Semantically Parse the future of your enterprise business

Foremost is that EinsteinDB uses a variant of multi-version concurrency control that interleaves OLTP transactions and actions without blocking OLAP queries. Another is that it uses an in-memory storage manager with lock-free data structures and flexible layouts that allows for fast execution of HTAP workloads. These design choices have already enabled us to implement support for some optimization actions in EinsteinDB. Over the last two and a half years we have designed, implemented, and deployed a distributed storage system for managing structured data at CloudKitchens, Uber, Netflix, AWS, Walmart, Salesforce, and others. EinsteinDB is designed to reliably scale to petabytes of data and thousands of machines via Lamport's distributed consensus algorithm which powers Raft, Multi-Raft, and Other BFTs utilized across the cryptocurrency and blockchain industry: privately and open source.

Vector Shape
Vector Shape
about

Beyond Lamport Clocks: Partitions, Transaction isolation levels, and serializability

To understand partitions, we need to know a little bit more about how EinsteinDB works. As with most ACID databases, EinsteinDB uses two-phase commit (2PC) and strict two-phase locking to ensure isolation and strong consistency. 2PC has been called the “anti-availability” protocol because all members must be up for it to work. Transactions in EinsteinDB will work as long as all of the touched groups have a quorum-elected leader and are on one side of the partition. This means that some transactions work perfectly and some will time out, but they are always consistent

Get Started
about
about