As time moves on, so does data. What was once big is now small, and what was once considered fast is now not even good enough. Ten years ago, Vertica found its place in the market with New Economy companies who were not just disrupting traditional industries but creating new ones. Social media exploded, wearables and sensor embedded devices became an everyday thing, AdTech became a word, and online gaming consumed us all.
In the early days, this meant hundreds of concurrent analyst queries, tens of compute nodes, and many terabytes of data. Twitter, AOL, Suunto, Delta DNA, Criteo, Taboola –these names didn’t sound familiar to most people just a few years ago, but they were core members of our Vertica community.
Small-scale analytics platforms reach their limits
Fast forward to today, when concurrency means thousands of analysts spanning multiple divisions and use cases, speed means sub-millisecond response time, and scale means hundreds of nodes in a single cluster without limits. These demands have disrupted traditional architectures and, combined with hybrid deployment expectations, are forcing some market vendors to explain their limitations.
Recently, it was brought to our attention that Amazon Redshift is limited to a maximum of 128 nodes, and even with 128 DS2.8XL’s, you’d find yourself with only 2 PBs of data. Just a few years ago, that was a lot of data. But today, analytics without limitations like this is the only option for companies like The Trade Desk, who is currently running two 320 node Vertica clusters on AWS serving 7 petabytes of data. When you’re running 10 million ad auctions every second, limits are not an option. (Be sure to ask the Trade Desk representatives about it at the Vertica Big Data Conference next month!)
And small-scale doesn’t translate to low cost
Another thing to consider is that small-scale doesn’t necessarily mean low cost. Because the cloud-only vendor Snowflake bundles underlying cloud infrastructure with software and charges a monthly fee, it’s predictable that the TCO is much higher than what most companies could achieve if they negotiated directly with the CSPs. Redshift has AWS to buffer its finances so costs are lower but they come with limits.
Be sure your analytics platform can scale as you grow
So, what’s the difference between the small-scale platforms like Redshift and Snowflake versus Vertica, which makes performance at unlimited scale a reality? Should companies who want to start small and test their business models focus on the small-scale options? It depends on how many times you want to change your analytics platforms. Vertica has worked with many companies who have graduated from small-scale platforms like Redshift and Snowflake and now are successfully meeting their Tier 1 business expectations and SLAs. But most of these companies regret the time and money they might have better utilized had they graduated to Vertica sooner.
For data analytics flexibility, focus on the architecture
Vertica in Eon Mode is built on an architecture that separates compute from storage, which enables exponentially larger cluster sizing combined with the ability to deploy sub clusters for departmental use cases, variable allocation of compute resources, and eliminating the overhead of continuous backup and restore.
Vertica’s MPP columnar architecture combined with the separation of compute and storage enables companies like The Trade Desk to scale their clusters and their communal storage beyond historical limits. There’s a clear correlation between the company’s financial success and their analytics model as cleverly pointed out by the Motley Fool in the recent “Could The Trade Desk Be a Millionaire-Maker Stock?”
Starting small shouldn’t limit your big data future
If you need to start small but want to be assured that you face no limits to future success, Vertica can help. Whether it’s Vertica by the Hour on AWS Marketplace, or Vertica Community Edition, or Subscription license per TB or per Node, we’ve got you covered. Need deep dive technical training on-demand, or hands-on certifications to build your Vertica skills? The Vertica Academy has it all at no cost because we believe knowledge is power and we want everyone to benefit from the power of Vertica. Today, Vertica is powering the data disruptors in every industry. Vertica is analytics without limits.
About the Author
Joy King
VP Product & GTM Strategy, Vertica
Joy leads Vertica’s Product and GTM strategy, including business strategy and execution, customer and partner education, product direction, and go-to-market initiatives. Joy has spent three decades in the software industry and is a passionate believer in the power of predictive analytics and the need for every company to become an industry disruptor.
