Rockset Shatters Operational Barriers for Real-Time Analytics
Serverless approach to real-time analytics now decouples storage from compute in the cloud to outperform operational model of Elasticsearch, Apache Druid and others
SAN MATEO, Calif. – July 28th, 2020 – Rockset, the real-time indexing database company, today announced new product features—including decoupling of storage from compute in the cloud, and separation of query-compute from ingest-compute—further eliminating the operational barriers in achieving real-time analytics at scale.
To stay ahead in the digital economy, companies need to personalize user experiences, build real-time decision systems and automate actions using sensor data. These types of applications demand both very low latency and complex analytical queries on a variety of data formats from different sources. Since real-time data is messy and bursty, the high operational cost and complexity in ingesting and querying real-time data at scale has been a major barrier for enterprises moving from batch to real-time.
Built by the creators of RocksDB, Rockset automatically builds Converged Indexes™ on any data—including structured, semi-structured, geographical and time series data—for high performance search and analytics at scale. It supports real-time SQL queries on semi-structured data, so for the first time ever, developers have the flexibility to build features that search, aggregate and join any type of data from any source on the fly. When combined with serverless operations, it is a game-changing approach to making real-time analytics fast, flexible and easy.
With this new release, Rockset decouples query-compute, ingest-compute and hot storage in the cloud, allowing developers to scale the exact resources needed in the cloud and seamlessly adapt to variability. Traditional systems suffer from compute contention during bursty ingest or spikes in queries, so data becomes stale and queries slow down—making it operationally challenging to enable real-time decisions. Separating query compute from ingest compute avoids resource contention, and as a result, data freshness is guaranteed and query performance is never compromised. Additionally, while new data continues to stream in, old data grows over time and is less frequently queried. Scaling storage independently from compute allows customers to seamlessly control their price-performance over the lifecycle of the dataset, simply by adjusting the compute allocation over time.
Rumble Wellness, a company that encourages a healthy lifestyle by rewarding users based on their daily step count, uses real-time data to increase user engagement. "Rockset is pure magic. We chose Rockset over Druid, because it requires no planning whatsoever in terms of indexes or scaling. In one hour, we were up and running, serving complex OLAP queries for our live leaderboards and dashboards at very high queries per second. As we grow in traffic, we can just ‘turn a knob’ and Rockset scales with us," said Yaron Levi, Chief Architect at Rumble Wellness.
With the new release:
- Data is served from hot storage, and automatically backed up on durable cloud storage, with per-GB pricing.
- Compute resources are scaled independently as required, with per-minute billing.
- The compute resources are used for ingesting data or querying data from hot storage. Scaling compute resources results in instant performance gains.
“First-generation real-time analytics solutions like Elasticsearch and Druid tightly coupled storage and compute because they were optimized for the datacenter era. But this leads to over-provisioning of resources and makes it difficult to manage costs at scale,” said Dhruba Borthakur, co-founder and CTO, Rockset. “On the other hand, scaling storage and compute independently in the cloud provides the benefits of improved scalability, availability and better price-performance ratios. Rockset’s cloud-native architecture and serverless technology already enables massive operational gains for customers, and these new features will further enable developers to achieve real-time analytics at scale.”
To learn more about Rockset’s new capabilities and how it stacks up against competitors, register for the upcoming Tech Talk—“Serverless Real-time Indexing: A Low Ops Alternative to Elasticsearch”—scheduled for Thursday, July 30 at 11:00 a.m. PDT / 2:00 p.m. EDT.
- Tech Talk: “Serverless Real-time Indexing: A Low Ops Alternative to Elasticsearch”
- Rockset website
- Rockset blog
- Rockset latest news
- Follow us on Twitter
- Join us on LinkedIn
Rockset is a real-time database in the cloud, built by a team of industry veterans with decades of experience in web-scale data management and distributed systems at companies including Facebook, Yahoo, Google, Oracle and VMware. Rockset is backed by Greylock Partners and Sequoia Capital. For more information, go to rockset.com or follow @RocksetCloud.