Macrometa was founded in 2017 by CEO Chetan Venkatesh and chief architect Durga Gokina. It’s a globally distributed high performance data infrastructure and edge compute service with ambitious plans to support a wide range of data models and modern event-driven programming. Macrometa will compete head on with hyperscale cloud companies, edge compute and data as a service providers. With a recent round of founding Macrometa is now expanding its edge compute points of presence, based in data centers around the world, and claims to have 175 so far.
It positions itself as a low latency stateful service on which developers can build and deploy event driven applications and functions globally. These apps can target data models including key value store, JSON-based document database, stream-based or even graph traversal.
The team has offices in San Francisco, Bulgaria, and India. In early October Macrometa announced $7M Seed financing, led by DNX Ventures, as it unveiled its first products.
Macrometa Global Data Network (GDN) is a fully managed data and compute service accessed via serverless API. The GDN combines a low latency, globally distributed streaming NoSQL database, pub/sub, stream data processing with a function and container runtime.
Macrometa has some hardcore computer science behind it, based as it is on research about conflict-free replicated data types (CRDTs), where a network can be replicated and updated concurrently, while maintaining consistency – “it’s conflict free!”. Riak by Basho was the first well known data store that supported CRDT, and was well known for its rock solid performance and consistency.
Go To Market
Macrometa has a two-pronged approach – aiming for bottoms up adoption by developers, but also enterprise sales led top down engagements with companies looking for high scale infrastructure for data-driven applications. Initial customers include major tech vendors, global retailers and telcos.
The market for distributed cloud-managed databases is extremely hot right now. Managed services are the new black.
As such the hyperscale cloud providers Amazon Web Services (AWS), Google Cloud Platform (GCP) and Microsoft Azure are all competitors, with their wide range of database and data storage offerings, ready made developer communities and customer bases. The hyperscale players not only offer managed services based on their datastores, but also third party services.
Given the developer mindshare of AWS in particular, Macrometa is kickstarting its own developer play by cloning the DynamoDB API, which makes for some cool demos and potentially easy on ramps for developers. At global scale however, the API cloning approach would be strained. Cloning the API is one thing, cloning how the system behaves at scale is quite another. As we’ve seen in other markets, you know, good enough API cloning may be good enough for some potential customers.
It’s also frankly a rather bold move to market on the basis that you offer higher performance and more regions than Amazon Web Services.
The points of presence argument may open some doors with customers given geopolitics, the pressure we’re seeing on the EU safe harbor agreement, etc. We’re seeing more interest and attention from government organizations, financial services and others in having more control over where they host their data, and with whom.
Macrometa is not only competing with AWS, GCP and Azure.
It argues it’s the future of edge compute, a space with strong incumbents. Fastly and Cloudflare, modern content delivery network (CDN) market leaders, are both aggressively investing in serverless edge computing, and both have strong engineering teams and excellent reputations in terms of performance. StackPath is pursuing a similar direction. Macrometa could partner with these companies, but competitive pressure will emerge pretty quickly.
In terms of distributed, managed database services targeted towards developers, the most obvious competitors are MongoDB’s Atlas and AWS DynamoDB. But other entrants such as FaunaDB are also aggressively courting modern serverless developers. Cockroach, Yugabyte and GCP’s Firebase are also in the mix.
See also Redis—another platform with significant developer mindshare—as Redis Labs is also looking at CRDTs as a means to support active-active geo distribution of Redis-based clusters.
*Disclaimer: Macrometa is a RedMonk client, but this research was not commissioned and all opinions were independently produced. AWS, Azure, Cloudflare, Fastly, Fauna, GCP, MongoDB, Redis Labs, and StackPath are also RedMonk clients.