Why Event Data at Scale Matters and How Scalyr Makes It Affordable

Event data is the lifeblood of managing a digital system, service or business. Until now it was difficult to keep and analyze the critical data emitted by these applications and infrastructure, especially as they migrate to a modern stack.


Scalyr believes it is essential to collect and retain all event data from applications, services, and infrastructure. Scalyr makes it affordable to keep all of your application and infrastructure event data.

Modern Stack Diagram

In modern stacks, the data that is generated increases exponentially with the move to microservices, containers and the cloud. Budgets are not keeping pace.

Why Does Keeping All of Your Event Data Matter?

For engineering, product, and operations teams, the real-time performance data of a system is essential for problem detection, isolation and correction, incident management, forensics, planning and optimization, trend analysis, behavior insights, and more. It feeds a variety of use cases as well as AI and ML engines. In addition to being a reactive tool, the use of this data can help improve performance, create competitive advantage, increase user satisfaction and engagement, and catalyze new product innovation.


Because of the cost and complexity of keeping event data and making it useful, users have had to discard or summarize data, often at the expense of context or the history needed to understand and solve problems. Scalyr has eliminated the need for these painful tradeoffs. With Scalyr, you can ingest as much data as you want, have that data be immediately useful and keep the data, in a usable form, for as long as you need – all at affordable prices.

Achieving The Breakthrough

In its most recent scale and performance breakthrough, Scalyr has been able to benchmark up to 200 terabytes of data input per day, breaking the previous industry record for SaaS log management by 2x. To make this possible, Scalyr’s architecture has some unique features including:

  • Separation of storage and compute, allowing both to scale independently
  • Efficient use of S3 with the cost benefits passed along to its customers including affordable long term retention
  • Ability to multiply cloud compute power dramatically to accelerate search queries to complete in one second or less 96% of the time
  • Columnar datastore with high speed search algorithms that make large quantities of high-cardinality data searchable quickly
  • Scalyr’s approach avoids the cost, complexity and delays incurred by indexing. Data is searchable immediately with no overhead.

The Results?

Engineers and SREs that use Scalyr can start with a large event dataset to understand history, patterns and look for answers. By asking questions and exploring unknowns in real-time, users of Scalyr can get answers quickly. It’s a different way of working and a better approach to log analytics.


To learn more about Scalyr’s 200 terabyte breakthrough and the unique features of Scalyr’s architecture, read our papers:

Download PDF

Event Data at Scale: The Foundation of a Modern Business



Download Report

To learn more about Scalyr’s record performance, read the Benchmark Report


Read the Blog

Read Steve Newman’s blog about all the elements that it takes to deliver breakthrough performance