The Value Pillars of Haevek’s Falcon Compute · Pillar 02: Cost
- More performance with less compute saves on monthly infrastructure bills.
- Tiered pricing is predictable and incentivizes Haevek to offer a more efficient compute platform rather than encouraging customers to consume more.
The problem
Problem
Spark, the dominant data compute platform, is free to use, but it requires massive memory and processor resources to be performant and available. Other options like Databricks use SaaS pricing; the provider makes more money when your pipelines are less efficient.
Bills increase steeply, sometimes unpredictably, blocking innovative initiatives and cutting into profitability.
Haevek drops infrastructure costs by 85%
Falcon is Haevek’s data compute platform. It is a drop-in replacement for Spark, Databricks, and other solutions that do large scale data induction and transformation. Falcon interacts with the same data sources and storage as other solutions.
Run your innovative pilots fearlessly
Haevek Falcon re-engineers the budgets of data compute in two ways: first, it uses predictable (and low) flat fee pricing that makes budgets for the software license the same month after month. Second, it is vastly more efficient and reduces the real cost of data platforms — compute.
Predictable flat fee pricing
Under the Falcon pricing model, the benefit from performance improvement flows to the customer automatically.
Falcon is built to be the most efficient distributed compute engine possible. We want to be a springboard to innovation for our customers. Consumption pricing does not support our goal. Under consumption pricing every efficiency improvement cuts into the service provider’s revenue. That does not get us to our goal.
Our tiered pricing model, based on cores running our Kubernetes based Falcon platform, allows customers confidence in their pricing and drives us to find ever more efficiency together.
Reducing the real cost center: compute
Haevek is a quantum leap in data compute platform performance. It uses Kubernetes to manage resources, including rapid scale to zero capabilities that actually get used. Falcon runs as compiled code with high performance memory management that prevents sawtooth spikes in compute consumption.
When businesses take our test flight, they frequently see the same work accomplished with one tenth the compute. The savings in compute completely pays for Falcon’s annual license in a matter of weeks.
Falcon customers pay a fixed annual fee based on the maximum number of concurrent cores used, starting at 64 cores for tier 1. Since Falcon can be run in a variety of conditions (on premises, cloud, air gapped, or edge), our licensing is separate from the cost of the compute resources.
Case study
Case study: 93% reduction in infrastructure
432 cores
on Databricks
32 cores
on Falcon
93%
reduction in compute infrastructure
A production real-time streaming ETL workload was processing multi-terabytes per day on Databricks using 432 cores for a single pipeline. The same pipeline, running the same transforms on the same data, required 32 cores on Falcon. Annual spend dropped from $464K to $66K for this job, an 85% reduction in total cost of ownership (TCO).
Note: The 93% figure reflects the compute footprint change. The 85% TCO figure is the net comparison including infrastructure and software license costs for Falcon and Databricks.
Through a Falcon Test Flight, customers can learn precisely how much Falcon can save them per workload.
In a recent benchmark, Falcon was 57 times faster than Apache NiFi and 14 times faster than Apache Spark on identical workloads.
Customer story · Health care company pipelines
Reducing costs for transformational projects
80%
reduction in the cost of scanning and extracting metadata
Why they migrate: higher performance and lower cost
The cost of compute, not licensing, is what drags down the possibilities that data initiatives offer us. By incentivizing greater efficiency, Haevek brings new possibilities to organizations who are inspired by their data.
“When we first met, what you told me sounded too good to be true. From the tests we’ve run, you’ve kept your word.”
Technical SME · Collins Aerospace
Learn more
Slash Your Databricks Compute Bill by over 80 Percent Without Migrating Your Data →
How much cost reduction can you see with Haevek?
While the typical workload sees a 75% decrease in cost, each project is different. To find out what your savings can be, we offer the 2-4 week Falcon Test Flight for you to pilot Falcon in your environment with your data. No data moved, no lock-in, training included. The benchmarks will show the performance increase and cost savings Falcon can provide.