Breaking down the details of Terabyte pricing
Earlier this week we released a new offering (available for download at the end of this month) in our warehousing portfolio: InfoSphere Warehouse 9.7.3. One of the major benefits with this release is the ability to purchase in Terabyte increments.
Since the announcement I have had many questions come up regarding how we price per Terabyte, What is charged/not charged and how this all works – and will now attempt to illustrate simply (famous last words) the mechanics here in this blog
1.) Terabyte pricing: Give me an overview of what this means
When you purchase enterprise software, there are a number of different licensing models. A few common models are user based pricing and processor value unit (PVU) pricing. User based pricing charges you based on the # of users you have using the software (essentially charged for your productivity) and PVU pricing charges based on your hardware/system (charged for your warehouse performance). Depending on your specific organization and information / data warehousing needs, either one of these can be an attractive option, yet this is usually short term.
As business analytics (and the underlying data warehouse) start to produce localized benefits, it is a common trend for new users (think departments) to request additional or net new system usage. With increased usage comes increased demand on the system (we will save the mixed workload discussion for another day) thus taxing the data warehouse’s performance. Both of these scenarios either increase the cost of the warehouse – or at least in the case of the 2nd one – decreases the overall system performance.
Our approach to Terabyte pricing (driven by customer requests) aligns to an organization’s business value of the analytical system. Clients of the Terabyte pricing option only pay for the data that they are using – independent of both users and system power. This means that companies new to business intelligence and data warehousing are able to accurately plan and account for system growth as they see fit. New users and system upgrades will no longer break the bank and organizations can monitor their growth as they further adopt analytics. Organizations already reaping the benefits of a data warehouse solution can potentially lower costs and add functionality depending on their overall data usage.
In the end the main point is that “You pay for what you use.”
2.) So what is actually considered ‘user data’ in the data
warehouse and is charged?
The definition of chargeable user data includes:
a.) User Tables: User Schema Tables, User Global Temporary Tables (including LOB/XML), Application Tool Data, etc…
b.) Derived Tables: Materialized Query Tables (MQTs), etc.
What is not charged:
Index Space and System Space
3.) How does compression factor in?
The Terabyte pricing size is calculated after compression
4.) How do organizations track their usage
With the release 9.7.3, we have included a new tool in the package that measures the used space. This tool illustrates a number of other useful statistics as well such as size consumed per table and compression usage per table.
5.)Has all InfoSphere Warehouse pricing moved to Terabyte
No – All existing pricing models remain in place – The Terabyte pricing option is truly an option.. We are not forcing anyone into it – Is up to the customer to determine what is the best alternative.
6.) How does pricing stack up against the competition
I’ll save this last one for another blog day – but have included the link to the summary report written by ITG – Suffice it to say, I’m very pleased with the findings.