Speedata provides a purpose‑built Analytics Processing Unit (APU) on a PCIe accelerator card that offloads compute‑intensive Apache Spark workloads to dedicated silicon, delivering up to 100× faster query execution without code changes. The APU integrates transparently via the Dash plugin, automatically routing eligible Spark operators to the accelerator while maintaining compatibility with standard 2U servers or OEM configurations. This hardware acceleration reduces compute time, data‑center space, power consumption, and total‑cost‑of‑ownership for large‑scale analytics and AI data‑preparation workloads.
Funding
$44M raised to dateRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
KDWCFounders
Product
Problem
Enterprises running large‑scale analytics and AI data‑preparation workloads on Apache Spark face severe performance bottlenecks because general‑purpose CPUs cannot keep up with the volume and speed required, leading to under‑utilized compute, missed SLAs, and high infrastructure costs.
Solution
Speedata offers a purpose‑built Analytics Processing Unit (APU) implemented as the C200 accelerator card. The APU offloads compute‑intensive Spark operations to silicon optimized for high‑bandwidth data processing, delivering up to 100× faster query execution without any code changes. Integration is handled by Speedata’s Dash plugin, which plugs into Spark’s Catalyst optimizer to automatically route suitable operators to the APU while delegating unsupported SQL elements back to the CPU. The accelerator is available as a PCIe card that can be installed in standard 2U servers or ordered pre‑configured with two cards, and it can also be sourced through OEM partners such as HPE. By moving analytics to dedicated hardware, customers achieve dramatic reductions in compute time, data‑center space, power consumption, and overall total‑cost‑of‑ownership.
Target Audience
Primary customers are data‑engineering teams and analytics platforms in large enterprises that run Apache Spark workloads for batch processing, ETL pipelines, and AI data preparation, as well as cloud and hyperscale providers seeking hardware‑level acceleration.
Features
- Dedicated APU silicon designed for analytics acceleration, providing up to 100× speedup on Spark batch ETL and AI data‑preparation workloads
- Transparent Spark integration via the Dash plugin that automatically identifies and offloads eligible operators to the accelerator
- PCIe‑form‑factor C200 accelerator card compatible with standard 2U servers and OEM‑provided configurations
- Pre‑configured server option with dual C200 cards for rapid deployment and out‑of‑the‑box performance gains
- Significant hardware efficiency gains: up to 94% space savings, 86% power reduction, and up to 91% capital cost reduction compared with GPU solutions
- Workload Analyzer tool that lets users benchmark their own Spark logs to quantify expected performance improvements