Todd Mostak
Apr 3, 2018

Announcing MapD Cloud: The First GPU-Accelerated Analytics Platform in the Cloud

Try HeavyIQ Conversational Analytics on 400 million tweets

Download HEAVY.AI Free, a full-featured version available for use at no cost.


Today we are incredibly excited to launch MapD Cloud. As we announced today, now anyone can start benefiting from the power of MapD’s GPU-accelerated analytics platform.

Users can literally be up and running on a 14-day, free trial of MapD Cloud in less than 60 seconds, with only an email and a few clicks. MapD Cloud is available via self-service Individual plans, as well as an Enterprise option for teams that require greater scale, High Availability, and direct access to their instance. Individual plans start at $150 per month, with a data allowance of 10 million rows, and range up to a 100 million row allowance.

It is fitting that we are launching this immediately following Nvidia’s 2018 GPU Technology Conference, since it was at this conference a year ago that we open sourced the MapD Core SQL Engine. In many ways we see the launch of MapD Cloud as the next major step toward our larger vision of giving everyone access to GPU-accelerated analytics.

Today’s announcement has been long in the making. While we have envisioned such a service since the early days of the company, the actual implementation was difficult, at scale, without widespread availability of GPUs in the public cloud. Back in 2016, when we made MapD generally available, it was far more difficult for a customer to set up a cluster of GPUs to run our software. Only a few hardware manufacturers carried GPU server SKUs, and they were scarce in the public clouds. Customers with pressing demand tended to find a way, but it was harder for mainstream adopters to justify procurement of seemingly exotic hardware to run software on the hunch that it might be game changing.

But over the last 2 years we’ve seen a dramatic rise of GPU computing adoption. Businesses and governments have purchased large quantities of GPU hardware to support burgeoning deep learning and machine learning initiatives. All of the major cloud platforms, including Amazon, Microsoft, Google and IBM, have added first-class support for GPU infrastructure.

At MapD we have taken advantage of both trends, partnering with hardware vendors to standardize on-premises deployments, and also launching on the AWS marketplace.

However, both adoption modes still involve a significant amount of friction when compared with today’s dominant software consumption model: SaaS.

The defining attribute of SaaS is that the entire experience of accessing and using the software is made frictionless by the vendor. Vendors typically know best how to deploy, support, and optimize their own software and the hardware it runs on. That is true for us at MapD, but our knowledge advantage is amplified when it comes to relatively unfamiliarity of GPU hardware. IT teams typically have less expertise in deploying this hardware compared to CPU-based systems, and GPUs can be comparatively expensive if they are not used at full capacity. MapD’s expertise in managing clusters of GPUs enables us to minimize end-user costs and improve efficiencies over time.

As the leading innovator in GPU-accelerated analytics, we are not stopping with what we’ve launched today. In the future, we will roll out more SaaS options and work to continuously increase adoption and to improve the user experience.

So, if you’ve found the logistics of GPU deployment to be a barrier to trying our platform, the MapD Cloud offering we are launching today was made for you. We invite you to spin up a trial today and discover what insights you can find beyond the reach of mainstream analytics tools.

Todd Mostak

Todd is the CTO and Co-founder of HEAVY.AI. Todd built the original prototype of HEAVY.AI after tiring of the inability of conventional tools to allow for interactive exploration of big datasets while conducting his Harvard graduate research on the role of Twitter in the Arab Spring. He then joined MIT as a research fellow focusing on GPU databases before turning the HEAVY.AI project into a startup.