What is most valuable?
IBM PureSystems or NPS has a patented NDAC or Netezza data accelerator card inside each of the SPUs (Snippet Processing Arrays). The SPUs act as individual processors that share part of a query. This helps in high-speed parallel processing of big data volumes. Querying a large data volume takes just a few seconds to minutes, if it is done right.
How has it helped my organization?
Analyzing years of data requires high processing power and storage. IBM PDA has exactly that. Years of processed data (tables) can be queried and retrieved based on management requirements. This can be done in minutes for analysis. This is extremely important in identifying trends for decision making in higher management, to serve customers better in today’s business environment.
What needs improvement?
It is a highly complicated architecture and only IBM engineers/support, or someone who worked on the hardware side of the system, can understand the system architecture completely.
This means that:
- Replacements can only be done by an IBM engineer
- Components are not generic and are not on the market
- Many of the systems are IBM patented
- Service support costs are high
Storage, although high, is limited depending on the rack configuration. For example, an N3001-80, which is an eight-rack Mako Server and also the top end model with the highest storage, can store a maximum of 384TB. Should your data exceed this limit, the storage cannot be extended.
It cannot be used for unprocessed data. The data has to be in a table format.
For how long have I used the solution?
We have been using the solution for five years.
What do I think about the stability of the solution?
We rarely had stability issues. You may expect an average of one outage every two months for a few minutes, if it is maintained properly. This is mostly due to “pollreplytimeout” errors of the SPU.
All the rack components are dual for redundancy. Almost all components have failsafe/backup, including the host that is configured in cluster for high availability. With this, the system is up and running all the time, or you can get it up at the earliest possible time in case of component failures/system down situations.
What do I think about the scalability of the solution?
We had scalability issues. Storage, although high, is limited depending on the rack configuration. For example, an N3001-80, which is an eight-rack Mako Server and also the top end model with the highest storage capacity, can store a maximum of 384TB. Should your data exceed this limit, the storage cannot be extended. You will have to purchase a new appliance.
How is customer service and technical support?
I would rate the technical support at 7/10.
Which solutions did we use previously?
We did not use a previous solution.
How was the initial setup?
IBM sets up the NPS and does the initial configurations. As an admin, you will have a completely configured system ready to work on.
What's my experience with pricing, setup cost, and licensing?
This is a multi-million dollar product. The software and hardware are both IBM patented.
Netezza SQL ('nzsql') runs only on PDA systems, and cannot be installed on other enterprise-class servers. Power consumption is high and these systems must have mandatory IBM support, due to the patented and exclusive IBM hardware. However, the benefits outweigh the costs if you have large data volumes that require fast analysis on a day-to-day basis.
Which other solutions did I evaluate?
We didn't really examine any alternatives. I started working directly on IBM PDA (Netezza) and I haven't had any chance to work on the competitors' products. I was not a decision maker in the selection of this product. This was a higher management decision.
A close competitor of IBM PDA is Apache Hadoop. As of now, no other product is as fast or as stable.
What other advice do I have?
You can definitely consider this appliance if you have:
- A large volume of processed data (tables) that are created on a daily basis
- Data that requires daily analysis, critical for decision making, and a budget to complement it.
This is one of the most stable and fastest data warehouse appliances available in the market today.
Disclosure: I am a real user, and this review is based on my own experience and opinions.
Jun 26 2017