Oracle SuperCluster Reviews

Our primary use case is as a converged infrastructure for the core database.

It has encouraged a simple and systematic capacity upgrade.

A valuable feature is the Exadata Storage appliance, optimized for database queries.

I would like to see control domain and zone management via a management console, similar to that of OVM.

Primary use case was a combination of a hardware refresh/consolidation of aging Sun M5000 equipment onto a half rack T5-8 SuperCluster. One of the core business drivers was the reduction of operational costs, while improving performance and (perceived) stability of the platforms.

The promise of consolidation is still a strong selling point, especially if the prospective client has applications that run on SPARC and has its data store in an Oracle Database.

There are many areas for improvement. For example, better guidance in terms of troubleshooting issues relating to ZFS, as well as better tools/diagnostics for monitoring that specific component, to better identify potential issues. Hardware monitoring via OEM 13.2 is not 100 percent, as Ops Center is still required in some instances.

There was an issue with zones rebooting due to a CRS bug (Version 12.1.0.2 at the time) which took Oracle support quite a few weeks to eventually identify.

While the general consensus was to maintain about 10 percent free space on ZFS file systems, it became clear that 30 percent free was actually required in order to maintain acceptable performance. Having to keep that headroom available impacted scalability to some extent.

As with Oracle Support in general, it depends on the engineer picking up the call. Some issues were resolved quickly and with detailed information, while others often dragged and went around in circles.

Previous solution was M5000-based and the switch was from a consolidation/hardware refresh point of view.

Initial setup had to be redone as there was some miscommunication and/or misunderstanding of requirements.

Unfortunately, I have not had the best experience with Oracle’s SuperCluster. Zones were rebooting at random due to Clusterware bugs, there were various ZFS issues resulting in performance decreases and node evictions and, I’m sad to say, we did not receive the strongest vendor support in resolving the issues.

Do due diligence on sizing and whether or not the solution is, in fact, fit for purpose.

Right now, it's just being built. It's for our financials system, so it'll be running Oracle Financials.

Because it's classified as an engineered system, it's all integrated, it's all supported by one vendor. We don't have to go to multiple vendors for support, it's all integrated under Oracle. So it's one place to call, one place to support it, which of course, has advantages in terms of not having to engage multiple vendors.

I think it's a highly integrated, powerful system that will support thousands of users, so that's why it was selected.

One of the limitations we see is its integration with other things, like other brands of storage or networks, or other brands of databases. So if it had capabilities to integrate really well with DB2 or SQL Server or Hitachi SANs, those sort of things, that would be a real benefit. Right now, it's fully supported only under all Oracle infrastructure.

It's not yet in production,so we're hoping it will be very stable. We have some Oracle Exadata which is very stable, so we're assuming it will be a stable solution.

Within the same infrastructure it's scalable but, as I said, its integration with other systems in our datacenter is where there's more of a challenge.

I haven't personally used tech support but I know there has been a lot of technical support sessions with Oracle on the build-out. I believe that's gone really well.

The old system was getting old. We were on Solaris 10, not even Solaris 11, so it was just a matter of needing a replacement of the infrastructure.

When selecting a vendor, important criteria include their having a significant marketplace share. You don't want some minor player. You want somebody who can provide really good technical support 24/7, has excellent response time, and the ability to provide replacement hardware quickly, with four-hour response times. So you're usually limited to the big spenders, like Oracle and IBM and HPE. Those are the main considerations: the support, technical expertise, and the ability to actually deliver on what they're selling.

I am not involved in the setup, but I think it's fairly specialized infrastructure because it's highly integrated, so it does require Oracle engineer expertise to do the build-out. It's not something you can put together yourself. It requires the vendor to be deeply involved in the setup and configuration and the install.

I'm not entirely sure who was on the short list. I think one of the critical requirements was that they be certified to support Oracle Databases, so that reduced the list a bit. And, obviously, Oracle being the makers of their own database system, was the prime candidate for that, in terms of the solution.

I would rate it as a seven out of 10. The reason it doesn't get a higher mark is because of the challenges in integrating it with other infrastructure in our datacenter. If it's not Oracle, it's more of a challenge to integrate with it. That is unfortunate because we have a heterogeneous infrastructure environment where we have many brands and models of infrastructure. Ideally, we would want to utilize everything and not have the solution be isolated and not able to talk to anything else. So interoperability is a really big benefit for us. And the SuperCluster has some limitations in that regard.

In terms of advice to a colleague, again, the important aspect is the interoperability with other brands of hardware, like for storage and for network and for databases and so on. If you have a need to connect to a lot of stuff, you should perhaps look at other systems as well, not just Oracle SuperCluster, because it's very Oracle focused.

• Software in Silicon

One of the clients has many applications with many Oracle standalone databases running on multiple traditional storage/server technologies. Those databases were integrated to fetch business data. Hence, there was database slowness; also, wasting storage and resources on some servers. The customer wanted to consolidate all his applications and databases into a single platform, i.e, Oracle SuperCluster. Upon migrating to OSC M7, the customer achieved significant performance, reduced cost for licensing and used the Exadata and SuperCluster features.

• Stability
• Optimization
• Sustainable overall performance
• Storage savings
• Additional M7 (Software in Silicon features)

I have used it for one year.

We encountered deployment issues with some initial configuration settings due to IP conflicts, lack of power capacity and required additional switches.

After migrating Oracle ERP and database, there was an issue with the printer drivers, and issues running some Arabic reports. After upgrading the firmware version for printers, and adding a workaround to avoid Arabic login and reports, all went smooth.

We faced issues with ZFS failover, due to misconfiguration. After correcting the settings, ZFS failover worked like expected.

I am fully satisfied with customer service; can relax for the next five years without any major upgrade to the environment.

There were three vendors involved. Overall, technical support was satisfactory.

The client was on traditional storage/server technologies; they switched because they wanted to have a complete Oracle consolidated solution.

Many teams' involvement was required for initial setup. It is a bit complex until the initial configuration and basic installation of the OSC M7 is done.

A vendor team implemented it; it was satisfactory.

There has been a significant reduction in the cost of licenses for software/hardware.

Oracle SuperCluster M7 is bit expensive. However, it comes with many sizes, so, a customer can choose the right size required for the business.

• Initial planning, such as network IP assignment (range), mandatory power/cooling/switch availability for a smooth installation

The most valuable feature is its ease of use and implementation.

The amount of consolidation you can do to a single platform is a benefit to my organization.

Out of the box, configuration could be made simpler. The new MiniCluster has the virtual assistant that configures the platform, whereas the SuperCluster does not.

I have not had any stability issues at all.

The problem with the scalability at this point is that, while it is field upgradable, in order to do a field upgrade at this point in time, the entire SuperCluster has to be rebuilt. They're working on a solution to have field upgrades available so that it doesn't require being rebuilt. However, at this point in time, that's the only option. Previously, that wasn't even an option. Once you bought the SuperCluster, it was stuck in the configuration that you bought it at for its entirety of its lifetime.

They have improved it, but they still have some work to do.

With technical support, when we open a ticket, it gets resolved in a timely fashion, and I am happy with the resolution, absolutely. One of the benefits of being with the SuperCluster is that it's all Oracle. The hardware’s Oracle, the OS is Oracle, the database software is Oracle, the application software is Oracle. Everything is all Oracle. You only have one throat to choke whenever it comes to support. They provide a one-call resolution process, where you open a service ticket and they will get together any of the back-office disciplines that they need to in the support groups.

Installation and deployment has some complexities. You have to know all of your networking requirements ahead of time. You have to know all of your host names. You have to know how the domains are going to be set up ahead of time. Then, you have to work that configuration worksheet with Oracle ACS services to have their engineers actually configure the install. Then, an ACS engineer has to come on-site and perform the install.

If you follow these steps, it's pretty straightforward. They've got the process down so that they lead you through step by step. It's still a process that would be easier to configure if most of the configuration could be done on the customer side without the intervention of ACS resources.

We were looking at the Exadata system, but the customer decided that they really needed to be able to put their applications with their database. Whenever you're dealing with a solution that isn't totally database-centric, then you need to look at something that can handle database as well as applications. The SuperCluster is the logical option because it does both.

The biggest hurdle is getting all of the power and networking requirements in place in front of the purchase. Once you purchase, expect at least 12 weeks to have everything in place before you can actually begin implementing the SuperCluster. The SuperCluster will arrive, but all of the other infrastructure components will have to be placed before that can happen.