EraDB, the database company that helps enterprises manage hyperscale, cloud-native workloads, today announced EraSearch, an Elasticsearch-compatible alternative for log management built on EraDB technology. EraSearch drastically reduces the complexity of ingesting, storing, and exploring large volumes of logs, and dramatically reduces the resources required to operate existing solutions. By improving on a traditional decoupled storage and compute architecture, EraSearch is able to provide lightning-fast access to logs in real-time, while ensuring that durable copies of data always live within object storage, such as on Amazon’s Simple Storage Service (S3), Google Cloud Storage (GCS), or Microsoft Azure Blob Storage. Available now, EraSearch cuts cloud hardware costs by up to 80 percent and operational costs by up to 75 percent. This new log management offering benefits companies of all sizes, whether they are ingesting gigabytes, terabytes, or petabytes of data every day.
“From a first principles perspective, the architecture of existing solutions like Elasticsearch is fundamentally incompatible with the current and future growth of log data, particularly in a cloud-native world. Built on decades-old technology, these tools were never designed to handle data at this scale. As a result, customers are now experiencing exorbitantly high costs and crushing operational toil for what should ostensibly be a simple storage and retrieval problem,” said Todd Persen, Co-Founder and CEO of EraDB. “By applying a modern, cloud-native architecture, we have solved these traditional limitations — by removing them entirely — and instead offer customers a low-cost, limitlessly scalable solution that can be run in any environment. Our goal is for our customers to effortlessly store and explore logs from any source, at any scale, at any level of complexity, and with EraSearch we have realized that vision.”
EraDB solves hard problems | Modernizes data storage for a cloud-native world
Within every organization data volumes are growing exponentially — going from millions to billions or trillions of high-cardinality data points generated daily. All companies deal with these growing volumes of data across departmental silos, mainframes, and legacy systems. According to IDC, the total amount of data created and consumed during 2020 reached 59 zettabytes and will continue to grow at a 26 percent CAGR through 20241. Companies rely on data-driven insights to remain competitive, but still struggle to effectively store data at scale. As a result, engineers spend an increasing amount of time fighting with data infrastructure, while those who need to analyze it are left waiting and unable to find the insights they need. In particular, logs have become business-critical: they are essential to troubleshooting systems, can be used in machine learning and predictive workflows, and are a cornerstone of auditing and compliance.
Many companies have chosen Elasticsearch for their log management workloads because of its perceived status as an affordable, open-source option. However, once these companies see their stored log volumes increase beyond a few terabytes, they discover that Elasticsearch is painfully inefficient and difficult to scale, which leads to exploding costs. For many companies these costs begin to eat significantly into their operating margins, while also requiring them to dedicate a significant portion of their headcount to manual Elasticsearch management.
Alongside EraDB Co-Founder and CEO Todd Persen, who was previously Co-Founder and CTO of InfluxData, a team of database, distributed systems, and machine learning experts have assembled to solve a core set of problems that can be applied to all time-series data management problems, most notably log management. Their first product, EraSearch, is a demonstration of the transformative change the EraDB architecture can bring to established, inefficient segments of the larger data landscape.
Founded in June 2019, the company raised $7M in January of 2020 from venture capital firms Foundation Capital, Array Ventures, Global Founders Capital, and world-class angel investors to build a product focused on making it easier to work with massive time-series datasets. The company currently has paying customers and a solid pipeline. EraDB pinpointed the log management problem as the first data problem to solve and will then apply its core design principles to other time-series problems.
- Decoupled Storage and Compute: Object storage is the perfect place to store high-volume log data, allowing the other tiers to remain efficiently stateless.
- Elastic Everything: Every tier can scale up or down corresponding with policy and demand, making it the perfect fit for deployments managed by Kubernetes.
- Dynamic Synchronized Caching: All data is stored in durable object storage, but is also always kept up-to-date in purely ephemeral caches, streams, and materialized views.
- Immediate Availability: Unlike Elasticsearch and other products, data is ready to be queried as soon as it is ingested.
- Faster Queries: Designed from the ground up for log management workloads, the highly tuned query pipeline is capable of outperforming Elasticsearch in most queries.
“At Foundation Capital, we understand that core database technology is hard to build and requires not just a re-factor, but a reboot. The market will continue to grow as every organization will need to use ridiculous amounts of data to drive insights and analytics to remain competitive,” said Joanne Chen, General Partner, Foundation Capital. “Todd and the EraDB team have unparalleled experience from the first generation of time-series database tech and they are uniquely positioned to usher in the next generation of core tech — EraSearch is just the start.”
“EraSearch offers organizations a compelling alternative for log management that is easy to deploy and manage, while giving them real-time views of their logs,” said Mike Leone, principal analyst, Enterprise Strategy Group. “Built on the EraDB platform by a team that understands solving time-series problems, businesses now can adopt a cloud-native architecture that is scalable and performant, at a fraction of the cost of other offerings available today.”
EraDB is a database architecture built on the core principles of decoupled storage and compute, a true zero-schema data representation, and flexible indexing powered by machine learning, all of which allow you to significantly reduce the size, cost, and complexity of your data while still giving you lightning fast queries across vast datasets. Our first verticalized product, EraSearch, incorporates these innovations into an Elasticsearch-compatible interface that is focused on serving the enterprise log management market. Find out more at www.era.co and www.erasearch.co.