ETL CompareETL Compare

Review · Read 13 October 2026

Speedata review: dedicated analytics hardware for Spark

ETL Compare staff · Figures checked September 2026 · Sourced from vendor docs, project pages and public benchmarks

In brief

Speedata pairs custom analytics silicon, the C200 Analytics Processing Unit (APU) card, with Dash, a Spark plugin that hooks the Catalyst optimizer, and states it needs zero code changes. It scores 2.61 out of 5 and ranks 9th of 9 on our lead ranking, because jobs have to move onto servers fitted with Speedata APU cards in your own data center, no price is published and stage-level detail is not documented.

2.61 / 5

ETL fit score (editorial assessment, 0-5)

ETL fit score (editorial assessment, 0-5)

Rank 9 of 9

Teams that can add dedicated analytics hardware on premises

Speedata at a glance
ApproachCustom analytics silicon: the C200 Analytics Processing Unit (APU) card plus Dash, a Spark plugin that hooks the Catalyst optimizer
Runs onSpeedata APU servers in your data center (pre-configured 2U server or OEM servers such as HPE); Amazon EMR, EKS and Databricks not mentioned on the product page
InstancesServers fitted with Speedata C200 APU cards
LicenseCommercial
PriceNot published; book a demo
Code changesNone stated; Dash Spark plugin (spark.plugins)
Read13 October 2026

Sources: speedata.io, Speedata product page, Speedata news, Speedata company page · Fetched 27 Sep 2026

How does it score on each criterion?

Score breakdown by criterion, with reasons
CriterionWeightScoreWhy
Adoption effort20%2.6Tied lowestSpeedata states Dash drops into existing Spark environments with zero code changes, but the job has to move onto servers fitted with Speedata APU cards, bought pre-configured or through OEM partners such as HPE.
Stage coverage20%2.6Lowest in setThe product page says Dash offloads compute-intensive work and that non-accelerable elements such as UDFs fall back to CPU; no operator list and no shuffle, spill or write detail are published.
Tuning and operating burden20%3.0A vendor product with a Workload Analyzer that reads your Spark logs; you run dedicated hardware in your own data center, and support terms are not published.
Platform and instance portability10%1.6Tied lowestDocumented for on-premises servers with C200 cards, plus a February 2026 sovereign-cloud partnership with Nebul; Amazon EMR, EKS, Databricks and Kubernetes are not mentioned on the product page.
Cost model transparency5%2.0Tied lowestNo pricing page and no public price for the software or the APU servers; buyers need a demo and a quote.
Published evidence15%3.0Vendor figures include 62.7x on production Spark workloads and 37 servers consolidated to 3; the site also cites a SIGMOD 2026 Best Paper whose Microsoft CoddSpeed benchmark features the APU. Setup detail for the vendor figures is not published.
Maturity and community10%2.6A commercial hardware vendor with Israeli founders; TechCrunch reported a $44M Series B in June 2025 (secondary source). Founding year is not stated and there is no open-source community.

What is Speedata?

Speedata is a commercial hardware vendor. Its product is the C200, a PCIe Analytics Processing Unit (APU) card, plus Dash, a Spark plugin configured through spark.plugins that hooks the Catalyst optimizer and offloads compute-intensive work to the card. Non-accelerable elements such as UDFs fall back to CPU, according to the product page.

The cards run in servers in your data center: a pre-configured 2U server from Speedata, or OEM servers such as HPE. Amazon EMR, EKS and Databricks are not mentioned on the product page. Speedata also offers a Workload Analyzer: you upload Spark logs, use a CLI, or compare against TPC-DS to estimate fit.

Sources: speedata.io, Speedata product page, Speedata company page · Fetched 27 Sep 2026

What does Speedata claim?

Vendor states

Up to 100x faster than CPU and GPU on accelerated workloads, and up to 90% lower TCO.

Source: Speedata product page · Fetched 27 Sep 2026

Vendor figure. Not measured by ETL Compare.

Vendor states

62.7x on production Spark workloads, and 37 servers consolidated to 3.

Source: speedata.io · Fetched 27 Sep 2026

Vendor figure. Not measured by ETL Compare.

These are Speedata's figures. We have not measured them, and setup detail for them is not published. The site also cites a SIGMOD 2026 Best Paper whose Microsoft CoddSpeed benchmark features the APU.

Where is it strong?

  • Speedata states Dash drops into existing Spark environments with zero code changes.
  • A Workload Analyzer that reads your own Spark logs before you commit.
  • Vendor-published evidence that includes an academic paper citation as well as customer figures.

What are the watch-outs?

  • Jobs move onto dedicated servers fitted with C200 cards, in your own data center (portability 1.6, tied lowest in set).
  • No pricing page and no public price for the software or the APU servers.
  • No operator list and no shuffle, spill or write detail published.
  • Founding year is not stated; TechCrunch reported a $44M Series B in June 2025 (secondary source).

Who should shortlist it?

Teams that run Spark on premises, can add dedicated analytics hardware and want to test a hardware approach with their own logs before buying.

Read 13 October 2026.

Which job stages does it address?

Spark ETL job stages and which stages each accelerator documents addressingA waterfall of five Spark ETL job stages (read, transform, shuffle, spill, write) with illustrative proportions, and below it a grid showing, for each accelerator, whether its own documentation says it addresses that stage.Anatomy of a Spark ETL jobRead: scan and decode filesTransform: filter, join, aggregateShuffle: write and fetch between stagesSpill: memory pressure pushes data to diskWrite: encode and commit outputIllustrative proportions, not measured data. Your own split comes from the Spark UI: see Profile a slow Spark job.
Stage coverage by accelerator, from vendor documentation
AcceleratorReadTransformShuffleSpillWrite
SpeedataPartialPartialPartialPartialNot statedNot statedNot statedNot statedNot statedNot stated
  • DocumentedDocumented: the vendor's own documentation says it addresses this stage
  • PartialPartial: indirect or experimental, or covered only by an end-to-end claim
  • Not statedNot stated: not found on the pages we reviewed

This shows what each vendor says, not what we measured. Sources are listed on each review.

Dash offloads compute-intensive work through the Catalyst optimizer; non-accelerable elements such as UDFs fall back to CPU; no operator list published.

Source: Speedata product page · Fetched 27 Sep 2026

Frequently asked questions

Does Speedata run on Amazon EMR or Databricks?

Amazon EMR, EKS and Databricks are not mentioned on the Speedata product page. It is documented for servers fitted with Speedata C200 APU cards in your data center.

Does Speedata need code changes?

Speedata states Dash drops into existing Spark environments with zero code changes, added as a Spark plugin through spark.plugins.

Compare with

Before you buy

Find your slowest stage with Profile a slow Spark job, then work out the money with Spark cost per job, explained.

Sources