EinsteinDB: Relativistic Causal Consistent Append-only Hybrid Database System.

exploring the combined compute/storage aspects of memristor arrays

we propose configurable hybrid data structures to improve the performance and lifetime of search intensive applications

big-data workloads

records can be stored in any of the six data structures: EinsteinDB, MilevaDB, VioletaBFT, FIDel, BerolinaSQL, Noether

memristor-based crossbar memory system

1M accesses per slice: TPC workload and 8M entries

About Us

An EinsteinDB and MilevaDB block of deferred columns (aka causet families or causet nets) can have several indexes and two index types are available: hash indexes and ordered indexes.

In EinsteinDB's Hybrid OLAP/OLTP workloads record accesses tend to be skewed. Some records are “lightlike” and accessed frequently (the working set), others are “timelike” and accessed infrequently, while “spacelike” records lie somewhere in between.

Logging and Offline Analysis
Forward Lightcone Algorithm
a single controller thread manages a set of worker threads each assigned to a single log partition
each partition contains consecutive time slices and the number of time slices in each partition is roughly equal.
Combined Compute and Storage
signal combination of MemCAM

we first propose EinsteinDB's and MilevaDB's MemCAM infra-as-a-service using classical CMOS, Silver Lake, R-NUMA, CC-NUMA; a configurable memristor-based content addressable memory (einsteinCAM).

More About Us

Our Mission

EinsteinDB provides automatic load-dependent scaling via AWS auto-scaling, which allows fully elastic provisioning of your storage and throughput needs.

The universal Envoy service persistence causal net mesh for distributed service connectivity

The EinsteinDB transaction service delegate all MilevaDB transaction connectivity and observability concerns to an out-of-process runtime: FIDel and Noether, both will be on the execution path of every request
another process running alongside our service process on the same underlying host
reduce the code that has to be written and maintained to build reliable architectures.
When Service Mesh first became mainstream around 2017, a few control planes were released by small and large organizations in order to support the first implementations of this new architectural pattern.
Greenfield-only
Universal and Kubernetes-Native

Envoy-Based: Powered by Envoy sidecar proxies, without exposing the complexity of Envoy itself.

latch-freedom for OLTP workloads

$0.05-0.10 per read IO/s

4kB page is around 175 sec

very fast response time and high throughput to mission critical applications

10% Horizontal Scaling OLAP/OLTP

standalone semi-relational knowledge-graph database that provides applications with extremely fast response time and very high throughput as required by many applications in a wide range of industries

Zero data loss

uses memory-optimized data layout and access methods for highly concurrent workloads

Our Services

We Offer Professional Solutions For Business

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