What needs improvement with Oracle Exadata?

The product must make more investments in object storage technologies. Six to seven years ago, the product had an SQL-based analytics feature that did not work well. The vendor must consider investing in it. The analytics features must be improved.

Patching must be simplified. We have various levels of patching in the solution. Most of them are online, split into various components. A single way of patching would be better.

Since the product is an appliance, it is very costly. And in the current age, people are cautious about spending this amount of money on any of these types of backend products. Some use cases are in real-time, where all other databases are much faster, but if you talk about the data warehouse, business intelligence, and all other perspectives in the transactional world, Oracle has to reduce the cost. Otherwise, a customer wouldn't want to continue this. If the same thing can be done at half or one-third of the cost, why would people stay with Oracle? Oracle Exadata would not have great value in front of a CFO. Other solutions can guard your data and address security concerns. Security, volumetrics, and so on are also provided by other databases, which are not that costly. Apart from Exadata, Oracle has other tools for business intelligence and other things, which they add on top of Exadata when they're selling a general license. For example, the Vertica database, an HP data warehouse. They have come up with their own analytic engine within the database, which gives an edge for the client to use the data analytics engine as a part of their database. Exadata does not have an analytic engine. Even MySQL has some statistical tools within it. If Exadata integrates analytical tools, it will be good for them.

There's room for improvement in terms of deployment, as it could be made faster and more user-friendly. I also have reservations about the cost, it is significantly high.

Oracle Exadata could improve by having faster data retrieval. We receive data at four or five seconds and want to reduce that number to one second.

The solution takes a lot of time to clone the environment. I would like to see some improvement in the cloning support or the time it takes on the storage side.

The handling of temp space has room for improvement. Oracle can make the handling of temp space better and faster. I would like to have locally attached temp space that would improve the solution.

One small area for improvement in Oracle Exadata is integration, particularly at the consolidated application level. Additional features I would like to see in Oracle Exadata in its next version include a cloud interface and GUI.

The problem resolution times with the solution are much higher. This information is based on our databases and our drive. The cost of the solution is high and can be improved.

The performance could be improved.

Exadata would be improved with single dashboard visibility.

Oracle Exadata could improve the monitoring system in the enterprise manager, it could be more user-friendly. In most Oracle tools there is a lot of functionality, and sometimes you need to do five or six clicks to find metrics, and sometimes it's a waste of time.

To use Oracle Exadata's Smart Scans and have it perform faster, I have to adapt the database, the statement, and the tables. Checking the Smart Scan issues is complicated, particularly troubleshooting the infinity band and the storage sales performance. In future releases, I would like to see more troubleshooting tools.

Oracle Exadata could improve by decreasing the price.

A room for improvement in Oracle Exadata is that it's not very easy to use in a microservices environment. It's not easy to split databases, and if this was easier to do in Oracle Exadata, it would make the solution better. What I'd like to see in the next release of Oracle Exadata is for it to become more modular, so you can use it in a context where the data layer is spread between many independent services.

We have experienced some issues with processing unstructured data on Exadata. This is an important requirement for our AIML based use case. Reactive analytics data can not be prepared easily in Oracle Exadata.

Oracle Exadata could improve the platform performance tuning should be easier, automated, and user-friendly. In a future release, I would like to have new analytical capabilities.

I liked Spark, but it was discontinued when Exadata L6 came back. I loved it, and I wish they would bring back Spark integration.

The setup is a little bit complex. We would like to see the installation part get easier. Also, this installation is only for the OEM. They need partners. Some liberty should be given to the customer when we talk about licensing. They should work to make it more flexible and give customers more options.

The technology in the on-premises version of Exadata is pretty much the same as the cloud edition. However, the cloud version is much more flexible when it comes to sizing and pricing.

I'd like to see a visualized console as part of the solution.

The solution could always be more stable and more reliable. The more they work on this, the better the product will be. That said, for the most part, right now, it's pretty good.

The improvement could be made on the hardware level as the habit in the industry is to go better and faster and larger with every iteration. From the software point of view, management point of view, it's okay right now. However, I don't understand why Exadata has no database nodes with SPARC processors. Oracle has SPARC servers that are on RISC processors and are more powerful processors than Intel processors. They never do Exadata with such processors on the database nodes level. However, they tested and it wasn't very useful. I would like to see Exadata with RISC processors on the database nodes if it's possible.

One of the things that we are facing during these few months is how you manage the people environment, and it's something that we have to explore to understand what is happening inside these machines. We are facing some problems with authentication. I believe that we are not familiar with the Oracle virtual machines. That's why we need to improve ourselves to better understand what is going on inside this environment. The integration with third-party applications regarding access management security could be better. I would like that to be included and be available the first time. Most of them, we need to buy later, and we would like to have these components when we're actually testing if this component. This is one thing that we are facing here in our environment. The integration becomes very difficult, and most of these features are not available the first time when you finalize the integrations.

There is one aspect to Exadata that I dislike, and that's the inconsistency with other databases. When you try to get Exadata to function with another type of database like SQL, or others, there should be reliable and consistent operation. When this is improved on, we should start to see more applications growing the market.

The replication of the database needs to avoid collision with the transactional databases. That's a very, very important aspect that needs to be looked at. It would be nice if we had access via mobile devices. To be able to have this information and the dashboards on cellphones or tablets or something like that would be great. It would make the solution similar to Tableau, for example, and other kinds of solutions. The initial setup process is very difficult and extremely complex.

A few improvements can be made such as better resource optimization, virtualization where you can see the virtual machine on top of the Exadata, the ability to limit CPUs, and reduced license costs. We need a monitoring tool which can in one place, where we can manage, monitor the entire Exadata components. Currently, we have multiple tools for different components to check and verify but one would be a benefit.

We've discovered that the solution is quite a complex product, which can make it difficult to sell. Snowflake, for example, is simpler and therefore an easy sell. The customization can sometimes be difficult to achieve.

Regarding the technical side of the machine itself, I don't see much that needs improvement. In terms of the kind of service and support that most of the clients need, it is huge investments. I would like to emphasize that the clients that use these technologies from Oracle must be well supported by the Oracle company. This is one thing that I would like to address. They could have better support. Additionally, the price for the Exadata is quite high. This is one thing that Oracle must think about. You can find the same features and the same performance that Oracle provides in other kinds of technology. So it depends on the client. If you want to use an Oracle engineered system, then you know that you have to pay. Otherwise, you'll need to buy more for performance, replications, and the availability of these kinds of things. But you don't want to pay a lot. You have another option that Oracle support calls Oracle ODA. With ODA, you don't have to use machines, but you have the same kind of features and key performances. However, you may have reduced options for scalability with these kinds of Oracle solutions compared to the engineering system like Exadata. After buying these, you have the support that you need to maintain all these environments. This is what I want to address. I will mention security. I know that there is a feature for security, but it is not included in the first purchase of this solution. That means if you need to increase the security, you need to buy the security feature which doesn't come by default on these solutions. As you may know, there are a lot of security problems all over the world with this kind of environment. Based on the fact that we are serving the government, we need to have security issues solved from the beginning and take care of security immediately. It would be better if Oracle could have some solutions that would bring us the confidence with their security at the outset. That's one thing I would like to address. I'm not saying that there is no security on this machine. There is good security on the version of Oracle which is running on these machines, it's very nice. But I'm saying this because I know that Oracle can do more than that and bring the substation to the clients.

There is room for improvement with the handling of the Temp IO, which is often used for JOIN statements. In my experience, rather than being handled by flash memory, it is being done using the hard drive. This slows it down a lot in certain cases. The reason this happens is that Temp IO often has a lower priority when compared to transactional operations such as Write Vault, Redo Vault, and Commit. This means that it can never get enough "attention" to be placed in flash memory. The entry-level pricing is too high for the smaller shops. Some additional built-in automation would be helpful. Some automation already exists in the software deployment process, but different components are handled differently, so more automation would be an improvement.

We still have to migrate to the latest version of Exadata, which we plan to do later this year or in the next year. Because of that, I am not sure we have anything that we would need to have added. I would need to consult our DBAs after we have migrated. They might find some issue that they would like addressed. But off the top of my head and because we are not on the newest version, it is not right to say the product needs something when it might already be there or has been updated. Exadata is practically a perfect solution for us as it stands. Because we are pretty satisfied with it, we have not rushed into the upgrade. I am not sure that we are fully utilizing the options that are currently on the table. For our contextual databases, it is the best option and we do not have any really an issue with it that needs improvement. We also need the product for other purposes. For analytics, we use Identity Two and we also need Microsoft Escrow Server for certain tasks. We accept some minor issues that I could identify if I discuss that with our DBAs. I do not think anything bothers us that much that we would need improvements. Of course, the price is the price, so it could always be less expensive. Maybe there are other considerations from the marketing side, but I do not deal with that. There are some issues with accounting where we really can not calculate return-on-investment. Exadata pays some fees for you so there is simplified billing, but that separates us from some evaluation of usage. Maybe Oracle could offer a solution for resolving that. Maybe a calculator or separate report that could help customers to find this data somehow. More clarity on this usage might affect how you estimate the workload of the storage and could really make clear what you get in return on using the product for the month.

The Oracle Support. I believe Oracle must improve its procedure to support the clients. The customer Ready Service must provide more use cases and benchmarks of their infrastructure to support client design decisions. Oracle must audit their partners regularly to guarantee they provide quality service even after been passed on partnership examination. Increase the FDP in the Southern Africa region can boost quality and competition on support service also increase product selling on these countries.

The management monitoring tools are quite important and an area that needs some improvement. The monitoring or consoles that are available should be available across the platform, and not only seen when logging onto the server. The availability of the monitoring should be responsive and available all of the time. I am planning to switch from Oracle Exadata to one of Microsoft's solutions, such as Synapse Analytics, to improve the performance. We have our Power BI and other parts in the cloud. The Exadata, being on-premises, creates problems at times because of the gateway. I recommend that Oracle come up with connectors that can be utilized by Oracle Exadata to convert the data that we have in Oracle to MySQL. We can extend the reach of Exadata to other toolsets.

It would be good if Exadata made some new features available regarding data retrieval and speed capacity functions.

General perception from many customers is the cost and maintenance of exadata database machine. Although Oracle recently introduced low cost storage and automated many features, still the cost and maintenance is a concern. Oracle should provide free of cost the Exadata software on the box to have more customers.

There is no room for improvement because everything that they could solve, they have. It has good storage optimization, hardware actualization, and internal communication network. They made improvements in whatever they could. The problem with the Oracle Exadata is that it is a dead-end of technology. The supplier's market is dying, so they have no future. For analytical purposes, companies are now using analytical databases. So they're using databases like MPP, multi-parallel processing databases, things like Greenplum, and Teradata. Exadata has no future. It is going into history. Five years ago, it was top technology. Now, they don't have a future. It's too expensive per terabyte. It's complex and hard to expand.

Try to diminish the logging service.

We have had issues with system restoration. If there is a system fail, the internal drives are useless. We need Exadata to integrate with that one, and to restore it back. But there should be something in place in case of system failure. One of the systems should have the facilities to troubleshoot from the other nodes to this note remotely. Or, even if not remotely, just to troubleshoot at least among the thrusters. There is no system collectively. Let us say this system failed; do we have any disks? Total disks, let's just say 100 disks. Five disks are failed. 100 disk measuring is not data top. If mirroring is that, like we can HRCMR. Let's say IBM CRM are CTC shadow image, REMC like that, the storage level they are giving the facilities inserted data mirroring, so immediately we can bring the DR. But internally, the Exadata doesn't have such features. Maybe they can think through that mirroring of collective system. Insert automatic measuring. If we have a collective recovery, let's say system-to-system, that DR solution, though they are giving a data guard, we have to set it up into the second system and the Data guard is a software solution. There is a really high skill required for a DBA to bring up the database. So a non-skilled person, like the storage engineer, can bring up for shadow images. HRCMI, HRCMR, like that. But in the case of Oracle, you must be really specialized to to bring up the data guard. In addition, they should insert automatic measuring. AM is, let's say, one table fail. Okay, you can recover it. Or one disk failed, then they can troubleshoot it. But automatic recovery, fine, but if we have a collective recovery, let's say system-to-system, that DR solution, though they are giving a data guard. We have to set it up into the second system and the Data guard is a software solution. There is a really high skill required for a DBA to bring up the database. For a non-skilled person, like the storage engineer can bring up for shadow images. HRCMI, HRCMR, like that. But in case of Oracle, you need to be really specialized to to bring up the data guard.

we do have a little trepidation with systemd as it does have a learning curve . Also changing to a binary logging format for us feels like retrograde motion , but sadly almost all linux varients have moved in this direction.

My biggest gripe has been patches which has dramatically improved since our initial Exadata was delivered (January 2011). The only issues we periodically experience are with non-default RPMs on the database nodes. These may fail during the pre-req check which means opening a SR with support. This has become the exception, not the norm so overall not much to complain about. The X2-2 used to experience frequent disk failures but now, that is a thing of the past.