HPE NonStop Reviews

• Rock solid reliablity
• 99.9% uptime
• Fault-tolerant transaction processing

Retail payments must be reliable with high uptime.

NonStop (formerly known as Tandem) in the Guardian OS could use another, friendlier macro language. The one provided (TACL) is a bit cumbersome - and there are not practical alternatives available.

I've used it for 12 years.

There was no issues with the deployment.

This is why stock markets use it for transaction processing. Stability is what NonStop Tandem is famous for.

You can't exceed the scalability available.

Excellent. Tech support is always, repeat always, a seasoned expert engineer.

I have worked on NonStop my entire career.

Relatively challenging, for example it was difficult to configure the communications.

We used a vendor team.

If you have a high transaction volumes, it can be justified. If your transaction volumes are ordinary, this may be way too much machine for your needs. Think five million retail transactions monthly, as an example.

If you are contemplating a fault-tolerant, high uptime transaction processing machine with above normal transactions volumes - to act as a transaction server with infrequent changes to code - and you simply must have bulletproof performance - this might be the ticket. And if it is, spend a little time checking into the flavors: original Guardian OS, or the Unix OSS. Primarily languages used are TAL, C, and, yes, Cobol.

The most valuable features for us are the security, availability, and hardware support.

Because we're able to get immediate support, this prevents downtime and directly impacts cost savings.

The hardware fails often, and I feel that HP really needs to pay more attention to this issue. We've talked to them about it.

We've been using it since 2000.

We've had no deployment problems.

It's quite a stable solution.

It scales to our needs.

The level of technical support depends on how helpful the first level is. Sometimes they understand our problems pretty well and escalate immediately when necessary. Other times they really don't know how critical something is so they ask a lot of questions and we have to send in log files, taking much more time than I think is necessary.

• Fault tolerance 
• 99.9999999% up-time

Very adaptable to modern systems with message-based/queue-based architecture.

More and continued modernization.

I've used it for over 30 years.

This is the standard for reliability and fault-resilience.

This is the standard for scalability.

The NonStop Support Center is very good, with an occasional bump in the road.

• IBM mainframes
• Bull mainframes 
• Unisys mainframes

Straightforward, with detailed installation manuals.

It was done in-house. I would advise you  to have two or three dry runs at implementation, documenting everything.

I would recommend you to perform very firm price negotiations. Also, get the best price point and buy-in from C-level execs ASAP.

• Protection against single failure. This makes it an ideal product for critical applications as found in banking (ATM control, EFT, web based banking etc), telecoms (administrative mobile phone support, switches), and any business where high availability is required.
• Linear expandability. HP NonStop is probably the only solution in the market that offers this really. Double the amount of CPU's, and double the amount of transactions can be processed. Double the machines link them at highspeed, and a double amount of transactions can be processed. This makes it ideal in fast growing markets like we had with EFT/ATM's and mobile telephony.
• Upward software compatibilty. Upgrading to a new operating system version means, in the worst case, recompilation of user software. Normally the software will function without applying any changes.
• Stability. The protection against single failure requires a very stable product, both hardware and software.

As I worked for various organizations in various branches, it's difficult to give a single example.

Normally in banking, replacing systems with bigger ones is a long term process with lots of carefully executed procedures. I remember a Dutch bank in the early 90's that measured in September an enormous increase in EFT transactions. Their capacity would fail in a couple of months at Christmas, and for sure they couldn't afford this. Beginning in November, a new system with doubled capacity was tested and installed thanks to the linear expandability and software compatibility.

Like on any system, hardware speed always could be increased.

I have been using the product since 1987. Since then it evolved as hardware from propriety CPU powered to industry standard Intel CPU powered. The operating system essentially stayed the same, although various add-ons were developed.

I've never encountered any serious issues blocking operations in this area.

I've never encountered any serious issues blocking operations in this area.

I've never encountered any serious issues blocking operations in this area.

Both customer service and technical support are of a high level, 8-9/10. However, this can vary based on the type of service agreement.

Both customer service and technical support are of a high level, 8-9/10. However, this can vary based on the type of service agreement. My experience is that the technical staff really knows where they're talking about.

As said, I worked for several organizations. Most of the time they already used HP NonStop or had already made the decision to use it.

Intial setup is straightforward. I was originally a software developer, but experienced no problems on system setups which I have done that several times. But, of course it is complex in the way that you need to understand what you are doing.

It is completely dependent on the system size, organization and branch. However, HP NonStop platforms can be used in smaller and bigger organizations.

It's dependant on the system size, organization and branch. If the operational support team can be minimalized, ROI may be high. But, for example, banks have procedures that require leveled support, and this increases the operational costs.

HP NonStop hardware previously known as Tandem NonStop with the Guardian NonStop Operating System. Hardware and Operating System are being used together to provide "NonStop" operations with protection against single failure of a hardware or software component. On earlier systems, CPU power and system memory was paired on a single board, but in the newer now-a-days systems all can be configured within normal limits.

Essentialy it is a propriety solution, although also a Unix variant operating system is available that runs in parallel with Guardian. It is a nearly perfect product, although it also has some restrictions and short comings.