Deploying a Big Data solution can be hard on a budget. When you factor in the cost of expensive analytical infrastructure, the numbers can really add up.
Case in point: Amazon's Redshift service provides cost-effective, flexible Big Data analytics services. Right? Unfortunately, over time, Redshift often proves to be among the most expensive solutions of all. Here's how the math works out:
Consider one of our recent customers that was looking to analyze 30 Terabytes of data with 24 servers, each with 16 virtual CPU cores, on US East.
After scoping their needs, the customer found that, although Redshift offers XL (2-core) and 8XL (16-core) servers, a 2-core server proved grossly inappropriate for fast analytics on Terabytes of data. Since they required real-time analytics, they realized that the 16-core 8XL nodes were the only realistic option.
At the time of this writing, Redshift On-Demand pricing (with no contractual commitment) costs $6.800/node/hour or $59,568/node/year. For a 24-server deployment, that costs over $1.4 Million per year!
Luckily, for steady-state production workloads, Amazon offers Reserved Pricing with an up-front commitment. With a 1-year commitment, including an upfront cost of $20,000/node/year, Redshift cost falls to $35,067/node/year. Or, with a 3 year commitment and including $8,000/node/year upfront, Redshift prices out to $15,989/node/year.
Adding it back up: for 24 servers, a 1-year Reserve pricing commitment will cost $841,613 per year. A 3-year commitment, arguably the most cost-effective option, will cost $383,739/year -- or $1.15 Million over the 3-year lifetime! And, if your analytic needs fall during that time (for example, due to the introduction of new technology), you're locked into an expensive contract.
Who wants to pay a Million dollars every year for analytics infrastructure alone? And, as data volumes grow, the price tag just increases over time.
At Calpont, we designed InfiniDB's perpetual per-core pricing model to accommodate situations like these, where:
- You realize that analytics will become a permanent part of your business going forward and want to avoid expensive per-year pricing.
- You want that your solution will scale with your data needs as they grow.
- You want to tailor your analytics solution to your hardware and enjoy falling unit economics on CPUs, memory and storage.
- You want to maintain control over your pricing and avoid being chained to vendors that can raise license fees at a whim.
Interested in pricing an InfiniDB cluster for your analytics project? Contact our Sales team. We'll provide a viable, competitive quote that won't burn a hole through your budget.