Skip to main content

Hourly Pricing: Vertica v. Redshift v. Snowflake v. Teradata v. Greenplum

  • May 11, 2018
  • 5 replies
  • 13 views

Forum|alt.badge.img+2

Now 9.1 is available for download or easy consumption on all the Cloud marketplaces.

Eon Mode is GA! including very exciting hourly pricing on AWS Marketplace: Click here

Now on to the interesting part! How does the Vertica Hourly pricing compare to the market? On the AWS cloud we have Redshift, Snowflake, Teradata, and Greenplum all available "on demand" by the node hour.

The best way to do this is to first split each offering between software and hardware costs. Next normalize each by the $ per CPU core. This is just a pricing comparison. Performance benchmarks are the next step to get a "MPG" rating but as we all know and proven by McKnight benchmark awhile back against Redshift, Vertica wins the performance per dollar comparison. This is because we optimize for performance and build features like Live Aggregate Projections and Flattened Tables. Snowflake a vendor which only survives on their SaaS per hour sales has zero incentives to make their software faster, they would prefer the customer to turn on more nodes!

First let's look at a simple chart comparing each vendors respective on demand per node hour price.

Not surprisingly you will find Snowflake is the cheapest followed by AWS Redshift.

Next, we look at the hourly price per node split between hardware and software.

Next, we look at the hourly price per vCPU split between hardware and software.

The incredible finding is Snowflake is the most expensive. They are actually more expensive than Teradata.

Vertica is in a perfect competitive spot. We are not cheapest. But we are much better priced than Snowflake, Teradata or Greenplum. We also have the best performance.

To make things even more interesting we can add the costs querying the data lake (parquet on S3). For the competitors this requires an additional tool. There is two choices offered by AWS, one is AWS Redshift Spectrum and two is AWS Athena. Athena is based upon Apache Presto. Both of these tools are query engines for the S3 data lake. They both charge for data processed per query.

AWS Athena/Spectrum processing 20 TBs per day $5 per TB resulting in $100 dollars of spend.

Finally as we discussed in enablement "AWS Private Offers" are available to customers who can commit to 1yr or 3yr usage paid upfront in return for a discount.

The details can be reviewed here:

Contact Mark Lyons for questions.
Mark.Lyons@Microfocus.com

5 replies

RonaldRong
Forum|alt.badge.img+1
  • Participating Frequently
  • May 11, 2018

Great information! Is there a downloadable format as well?


Forum|alt.badge.img+2
  • Author
  • Participating Frequently
  • May 11, 2018

Hi Ronald, We will be posting on enablement portal today and sharing via email.

More information about private offers is here


RonaldRong
Forum|alt.badge.img+1
  • Participating Frequently
  • May 15, 2018

I second this! I have heard the same comment from other customers in EMEA. The bigger customers are not waiting for a managed service but the smaller companies that cannot afford to employ many IT resources would prefer an option like a managed service. Currently we do have some partners 'selling' DBA skills but that is often not a cheap option either.


Jim_Knicely
Forum|alt.badge.img+2
  • Participating Frequently
  • May 15, 2018

Didn't we try this with Vertica on Demand?


Forum|alt.badge.img+2
  • Author
  • Participating Frequently
  • May 17, 2018

Please let me know any prospects or existing customers considering moving off of Vertica because of SaaS.

The paid hourly in the AWS MP should help if the objections are pricing but won't help if they want SaaS.