Executive Vice President & Chief Corporate Architect, Oracle Corporation
Oracle Enterprise Manager 13c Release 5
Edward Screven, chief corporate architect, Oracle, and Mughees A. Minhas, vice president of product management, Oracle, discuss product roadmap and enhancements. Two demos are included on the new patch automation UI and custom dashboards.
Find out more: https://oracle.com/manageability
About Edward Screven
Edward Screven, Oracle's chief corporate architect, has been promoting the company's MySQL HeatWave service, which he described as one of the fastest-growing OCI services since its introduction in October 2020. In a January 2023 presentation, Screven announced new machine learning capabilities for MySQL HeatWave, stating that model building, hyperparameter selection, training, and inference all occur within the database. He said the service allows users to perform machine learning without needing to be a data scientist and that the marginal cost of HeatWave ML is zero because it is included. Screven contrasted the service with competitors, saying that Redshift's AutoML exposes hyperparameters that shift the tuning burden to the user, while MySQL HeatWave's AutoML requires no such configuration. Screven has also discussed Oracle's internal use of its own cloud infrastructure. In a December 2021 presentation, he said Oracle runs its entire business on its own cloud, generating $41 billion in revenue annually. He stated that the company's migration to Oracle Cloud Infrastructure reduced compute requirements by 50%, increased performance by 30%, and reduced system provisioning from weeks to minutes. Additionally, Screven has spoken about Oracle Enterprise Manager 13c Release 5, which he said is designed for customers with both on-premise and cloud infrastructure, allowing them to manage both environments with a single tool. He cited a model from Peak Solutions that estimated a 60% reduction in operating costs for a hypothetical customer using Enterprise Manager.
Transcript (34 segments)
E
Edward Screven0:06
Hi, I'm Edward Screven, I'm Oracle's Chief Corporate Architect. I'm here today to talk to you about Oracle Enterprise Manager 13c Release 5, which we're announcing today.
Now, many enterprises of course are transitioning to cloud. That means moving from almost an entirely on-premise infrastructure to cloud. That of course takes time though, and not every customer is going to move completely to cloud. And so what that means is that you're going to have both on-premise and cloud infrastructure that you have to manage.
That mixed environment means that you need tools that span both on-premise and cloud. So your system management requirements are changing. You need to be able to manage systems and targets both on-premise and in the cloud using exactly the same tools. Because if you don't have exactly the same tools, what's going to happen is your labor expense is going to be higher. It's a lot more work, it's a lot more expensive because you're using different tools, right? And you're not going to be able to respond as quickly. You're going to make more mistakes. So having one single tool that can manage both on-premise and cloud together at the same time is critical.
Now, if you look at the way cloud applications operate, the pace of change in cloud is much higher than it traditionally is on-premise. And that means that you have both an opportunity and a challenge. The opportunity is that you can evolve your applications to satisfy business requirements much more quickly than you can using traditional IT approaches. The challenge is that you have to keep up with that change. You have to be able to take advantage of that pace. And that means taking advantage of that pace and responding to that pace both on-premise and in the cloud at the same time.
When you move to cloud, there are new tools to use. When you move to cloud, there's a wider variety of systems that you want to integrate with. So that means extensibility of your system management tools is also critical.
Oracle Enterprise Manager has been the premier tool for managing the Oracle stack for many, many years. It has some very important core capabilities. For one thing, it scales very broadly. You can manage literally millions of targets. Oracle, we've used Enterprise Manager internally. Our biggest deployment of Enterprise Manager is 2.5 million targets. Not only does it manage all those targets, it gives you tools to make sense of all those targets. I mean, it really doesn't matter if the tool scales to millions of targets if a human being can't deal with all those targets. So Enterprise Manager, through things like event coalescing or templates or dynamic groups, they make it possible for human beings to make sense of all those targets.
As I mentioned, Oracle Enterprise Manager is the premier tool for managing the Oracle stack. So you get deep management all the way from operating system through virtualization to database to middle tier applications, the entire stack. You can also use Enterprise Manager to manage engineered systems like Exadata or Database Appliance. It is by far the leader in managing Oracle Database. You get full lifecycle management, of course. But think about all of the decades of expertise and experience and best practices that Oracle has learned with the help of our customers on how to manage Oracle Database. If you want to get the most out of your Oracle Database in terms of performance, in terms of reliability, in terms of compliance, in terms of security, Oracle Enterprise Manager is really the only tool to use.
Now, today we're announcing Oracle Enterprise Manager 13.5. Oracle Enterprise Manager has a lot of new features, of course. I'm going to focus on features though that fall into those three buckets that I mentioned earlier: hybrid cloud management, automating operations, being extensible. Those three key categories of functionality that matter a lot more now in this hybrid IT environment we find ourselves in.
So with Oracle Enterprise Manager 13.5, you can manage your entire hybrid fleet. You can manage targets both on-premise and in Oracle Cloud and running in Oracle Cloud Customer. By the way, you can run Enterprise Manager itself either on-premise or in Oracle Cloud. It has direct support for Autonomous Database and OCI, for Exadata Cloud Service and OCI, and of course, as I mentioned, Cloud Customer running in your data center.
It lets you have something called hybrid database as a service. So if you're managing databases and you're managing environments in both on-premise and in the cloud, you can use Enterprise Manager to create a hybrid database as a service. So that means your application developers, your DBAs, can create database instances in any of those places, wherever it makes the most sense for your application, for your business.
Now, as you migrate to OCI, of course you want to move the data from your on-premise database instances into cloud. Well, Oracle Enterprise Manager 13.5 has something we call Database Migration Workbench. You tell it the source on-premise, you tell it the target in the cloud, and it figures out what's the best way to move the data. If you're going to use something like RMAN copy or you're going to use Data Pump or something else, it actually then manages the movement of the data and then validates that the data was moved correctly.
You can also use something called CloudBridge to connect Enterprise Manager with OCI. And so what that does is that takes the Enterprise Manager repository, copies it up into OCI, into OCI Object Store, and then you can use OCI services like Log Analytics, right, to actually analyze that data and gain insight into what's happening in your environments.
Before, I mentioned how cloud changes the importance of automation. Because the pace of operations in cloud is very different than traditional on-premise, automation becomes even more important than it was before. One thing that's very important, of course, is to be able to deal with all of these targets in all of these environments. You need to be smart about how you actually present event information to the user. So just to give you an example, if a host goes down, there's going to be a large number of components that are now down, right? It does you no good to see, you know, event down notification for the hardware, for the operating system, for the database, for the middle tier. You want all those compressed into one single notification, and that's exactly what smart event compression does.
Enterprise Manager 13.5 also has something called dynamic runbooks. So your DBAs, your administrators, can take everything they know about how to correctly diagnose problems, how to optimize, you know, performance problems, they can encode those into actionable rules. These dynamic runbooks that can be invoked in response to Enterprise Manager events.
Enterprise Manager 13.5 has something called automatic workload analysis. So what it does is it monitors the performance of systems over time and determines that there's been some variance, and it can then alert your DBAs to know that there's an issue that they should examine now, before it becomes a problem in the applications using the database.
Oracle Enterprise Manager has helped you manage your compliance status, and we've extended that compliance framework to add new targets and new standards. So for example, we can now do PCI DSS and HIPAA for Oracle Linux 7 or Linux 8. And we also have a new fleet management UI because you have this wide variety of systems across your on-premise enterprise, across your cloud, across things like dedicated regions or Cloud Customer. You need a very effective user interface to manage that entire fleet, to keep that entire fleet patched, to keep that entire fleet running smoothly.
As you move into cloud, of course there are new kinds of tools that you may begin to use. So for example, you may use Ansible, you may use Terraform. What we've done is we've added REST APIs to Enterprise Manager. And so what that means is that you can make Enterprise Manager be part of whatever flows you're creating in those tools. You know, if you're using Puppet or Chef, same thing. You can use those REST APIs, you know, for monitoring to query Enterprise Manager, to cause Enterprise Manager to create deployments, right? You can use Terraform templates to, when you go to instantiate some kind of service that you've built, Enterprise Manager can be part of that flow.
Many of our customers have multiple Enterprise Manager deployments. We do inside Oracle. You can now federate those Enterprise Managers so you can get one single pane of glass that lets you monitor what's happening across those Enterprise Manager domains. Finally, you can use our new dashboarding capability to create custom dashboards that give you rich visualizations of data monitored by Enterprise Manager. And not only can you use our dashboard tool to do that, we have a Grafana app. And along with that Grafana app, we have several built-in views of Enterprise Manager data through Grafana.
Okay, and one final thing is, okay, that all sounds good. You've met a lot of great requirements for cloud, you know, help you manage on-premise, help you manage in the cloud, help you manage both together, which is critically important. But okay, you know, is there some way to quantify that? Well, one third party did quantify that. It's an outfit called Peak Solutions. And what they did is they went and they interviewed many Oracle customers, actually customers across the world, North America, Asia, Europe. They talked to those customers, they understood how those customers used Oracle products, they learned about the nature of those customers, and they created a model, right? They created a model based on all of those conversations with all those customers, and they applied it to a hypothetical Oracle customer, a customer with 30,000 employees, you know, to give you some idea of the scale. In their estimation, without Enterprise Manager, it would cost $20 million, almost $20 million to manage that Oracle environment. With Enterprise Manager, it costs less than $8 million. An $11.9 million savings. 60% reduction in operating costs. 60% less cost because I need to employ less people. 60% less cost because I don't need to buy so many third-party tools. But I really think that actually underestimates the actual savings from using Enterprise Manager. I mean, the real savings, right, is the opportunity cost, right, that not having a tool like Enterprise Manager forces upon you. Using Enterprise Manager, you can respond to new requirements much more quickly. Using Enterprise Manager, you can migrate to cloud much more quickly. Using Enterprise Manager, you can manage that hybrid environment between on-premise IT and cloud. I think that's just critical. And so to tell you more about it, I'd like to turn it over to McGee's, Vice President of Enterprise Cloud Manageability.
M
McGee11:36
Thank you. So I'm going to drill down and provide some more details on this latest release of Enterprise Manager. So let's start with hybrid cloud management. You can manage your software assets with Enterprise Manager no matter where they are deployed: on-premises, Oracle Cloud, or on third-party clouds using the EM agent. Now, if you are running your databases in Oracle Cloud Infrastructure or OCI, we have new target types that are cloud-aware, such as targets for Autonomous Database, targets for Exadata Cloud Service, or Exadata Cloud Customer. So what that means is that for lifecycle management operations like startup, shutdown, backup, and so on, we actually call the native OCI APIs to perform these operations and do not do it directly with Enterprise Manager. So you can manage all these environments using the tool you have used for years with the support of hybrid cloud management that we provide. And you don't need to learn new skills. Your existing best practices, scripts, jobs can be applied to both environments equally.
We also give you choices for EM deployment: on-premises or in the cloud. You are obviously very familiar with how to deploy EM on-premises. For the cloud, we now provide a VM image with HA and disaster recovery configuration on the OCI Marketplace. So instead of deploying on your own data center, you can install EM and get it up and running with multiple OMS and high availability configuration in really about an hour or so with just a few clicks. This EM can be used to manage not just cloud assets but also assets running on-premises, and you would do it by setting up FastConnect or VPN.
Database as a Service. The next feature has been a part of Oracle Enterprise Manager's Cloud Management Pack for Oracle Database for quite some time. We are now extending it to support hybrid cloud environments as well. You can now use this capability to do self-service provisioning of new databases on Oracle Cloud Compute VM or bare metal, as well as on Exadata Cloud. Once you have established connectivity between on-premises and OCI, you can use the same REST APIs to provision and manage databases on-premises and in the cloud.
Moving on, in the area of data movement, we are now providing complete guided migrations using Database Migration Workbench to move databases between on-premises and OCI. This workbench helps you plan the migration, perform prerequisite checks to make sure that the migration can occur successfully. It then selects the best data moving technology based on the source and target choices you've made, such as database versions, type of database, for example, Autonomous Database, and so on. And then, of course, it executes the actual migration plan. For post-migration, it actually uses the SQL Performance Analyzer feature to make sure your workload on the migrated platform will perform well. So if any performance regressions are detected, it reports them to you so you can take proactive actions before opening up the database for end users. The Migration Planner feature, which also has been in EM for quite some time, now supports OCI as a target destination, including Exadata Cloud Service and Exadata Cloud Customer platforms.
So CloudBridge is a new feature that connects EM to Oracle Cloud. If you want to use an OCI cloud service and have it operate on EM repository data or data from EM targets, you don't need to establish connectivity between OCI to all the different targets running on-premises. The CloudBridge can extract data from these services and upload it on OCI Object Store through an automated ETL process, from where these services can consume them. To begin with, OCI Logging Analytics service and the Operations Insights service will use the EM CloudBridge to streamline interaction with on-prem data. The data extracted by CloudBridge includes EM metric data, configuration data from EM repository and from targets. It collects SQL performance data as well as this metrics data.
We have three key enhancements in the area of ops automation as it relates to observability modernization. First, we have dynamic runbooks. IT operations teams typically have standard procedures called runbooks to diagnose and resolve incidents. In this release, we allow these teams to create these runbook procedures directly inside EM. This allows teams to encapsulate EM and the operational expertise in diagnosing and resolving incidents. When an incident is raised, IT ops teams have easy access to the appropriate runbook and can quickly execute the runbook procedures. These procedures can dynamically invoke a metric chart from EM or execute a SQL or OS command. This promotes more consistent and faster incident resolution, while hopefully reducing your overall mean time to recover.
We also have smart event compression, which automatically compresses related events. So let's say if there are availability events, right, for example, if there are a number of availability events that happen within a narrow time window on certain target types, we can automatically compress them. So IT ops can also implement their own custom event compression as well. Together, these reduce the overall set of incidents that IT operations teams need to manage. Finally, I should mention that our Incident Manager UI has been redesigned with a simplified UI that highlights critical information and provides quick access to common user actions, thus providing a more modern user experience within managing incidents.
In the area of performance management, we have a very cool new feature called Automatic Workload Analysis. As the name suggests, EM constantly compares your current database performance with a baseline workload, and any divergence from this baseline is highlighted on the database home page. Examples of changes highlighted are, for example, regressed SQL statements and the quantified impact on the workload. So we can, for example, we can tell you that these three SQL statements out of 50,000 regressed and their impact on the workload is 7.2%, you know, something like that. Reasons for regression: there were plan changes, perhaps increased CPU, IO consumption, things like that. Also, if there are new SQL, added SQL that was not seen before in the baseline workload, we highlight those as well. So if you're a DBA, I'm sure this feature would be of immense interest to you.
Next, we have EM Warehouse. So this feature allows you to continuously extract EM data using a CloudBridge that I mentioned earlier and puts it in a database running on Oracle Cloud. You can now use rich EM metric and configuration and target data to do all kinds of analysis without impacting EM's core function, which is fundamentally monitoring and management. So on this data lake, you can do performance trending, you can do capacity planning, or any other kind of analysis using the tool of your choice.
Moving on, we have new UI for our database maintenance fleet feature. In just a minute, I will actually show you a demo of this feature where I will highlight its key capabilities. In the area of compliance and security tracking, we have three new things. First, we have out-of-the-box compliance standards for CIS for Oracle Database 19c, and STIG, PCI DSS, and HIPAA for Oracle Linux 7 and Oracle Linux 8. We now integrate with Oracle Autonomous Health Framework, which has been a long-standing ask from our customers, and particularly as it relates to Exadata environments. So now you can fully secure your engineered system components as well. We also have a new detailed HTML compliance report that you can review or email to your compliance colleagues for prescribed action.
Now let's go to the first demo of the new UI for fleet maintenance. Okay, so we start at the dashboard that helps you identify missing patches as well as configuration drifts. Here you can see there are 26 targets that are missing patches. Let's take a look at how we can patch these targets at scale using the new UI. So you go to the Targets menu, you select Databases, and once you're on the database page, you go under Admin and you pick Fleet Maintenance. So this launches the new UI. So the first step here is I need to select the gold image that I will use to deploy these patches. So I'm going to pick my image. Let me pick Database 19c as the gold image. I will next pick the type of database. Okay, let's say I'm going to only apply to cluster databases. And the third part is operations. We support both updates as well as upgrade operation. For the purpose of this demo, I'll pick update. Having done that, we can now select which database you want to apply to. So let me just pick a few. I pick HRDB1, FinanceDB1, AppDB1, FDB2. Databases selected, I click next. And here I will specify my different options. So I'm going to do both things. I will do software deployment part of it as well as update database. So I will provide the destination directory for my Oracle Home here, and I will schedule it to run immediately because this is an online operation. When it comes to the update part, since it's offline, I'm going to schedule it for later tonight at 6 PM. For credentials, let me just use preferred credentials to make it easy. And under options, I'm going to pick rolling mode as yes because I selected cluster databases on the previous screen. I can actually patch my database one node at a time, which means that the entire patching process is going to be entirely online. We also give you the option to add any custom script either in the pre or post deploy phase, or pre and post update phase. So I can, for example, pick here the force deploy script that will run post deploy, and this is my LDAP configuration script. So let's configure my LDAP for me. So this is a fairly handy thing where I can insert any custom script I want. With this, I click on next, and here's my summary screen. So I will be patching four databases, one of them has a standby configuration, as you can see. Before I hit submit, I'm going to hit the validate button, and this does the prerequisite checks and validation process to make sure everything is kosher and things will proceed successfully when the actual patching begins. It's also a good idea to do pre-validation before you hit submit. So things seem to be fine for us. You close it, and then before I submit, there's another handy thing where we let you provide us a prefix name for the job, so easy for you to identify the job in case something goes wrong and you need to do some triaging. So I'll just enter, let's say, my patch demo April, so I know when I submitted it, and I hit submit, and that's it. So it's pretty much as simple as that. This patching process will go on. So next I'm going to show is, you know, if you want to diagnose and troubleshoot some issues, I will show you an older run that we have run and see where you would go to diagnose those things if you have any issues with the patching process. So here we have a rich UI where EM can display all the relevant logs on the UI itself, and it also supports actions like, you know, debug, retry, ignore, etc., which are very useful when you're troubleshooting issues during patching or upgrade. So this concludes the demo for fleet maintenance UI, and hopefully it's something that you would find very useful in this new release.
So in the area of extensibility, we have a number of noteworthy enhancements as well. First, you can use REST APIs to execute management operations in Enterprise Manager. This means that you can invoke EM from your automation tool of choice, such as Ansible, Chef, Puppet, Terraform, anything that can call REST APIs and perform management operations like agent deployment, job executions, credential management, and so on. In the past, you've had to use EMCLI to do it, but now we have support in EM with REST APIs to do the exact same things. So that should be also a fairly handy and useful feature. The next feature I want to mention is dashboards. EM now has a dashboard framework that allows you to create a custom dashboard so you can visualize data as you wish. I will demo this feature shortly. We also have added new dashboards to the EM app for Grafana. This is a new dashboard, there's a new dashboard for WebLogic Server and another for JVM performance reports too. For EM Mobile App, which we also released last year, we now provide notification based on both SMS and email. Finally, if you have multiple EM sites in your environment, you end up with data silos and you don't truly have an enterprise-wide view. With the EM Federation, we can aggregate monitoring data from multiple EM sites and show you the monitoring status of all your targets across all your environments from a single dashboard.
So now let's take a quick look at the dashboard demo. So let's start by showing you how you access the new dashboard. So you go to the Enterprise menu, and here you see the link for Dashboard. So I click on that. And on this dashboards page, I have two options: I can create a new dashboard or I can use one of the existing dashboards we ship out of the box in this new release. Let's try and create a new dashboard. So here's my dashboard canvas. On the right are a set of widgets that I can choose from. So for this demo, let me start by maybe Target Status Demo, make it a little bit bigger. I'm going to add to it Incident widget as well, make it bigger as well, and perhaps add Compliance to it and Problems. And then, you know, that pretty much is how simple it is. We give you a library of widgets, you select from them, you size them as you wish, you pick the data you want to see, and then you give it a handy name. I call it My Enterprise Dashboard, and I click on Done, and I'm done. So my dashboard is created. So that really is how simple it is for you to create dashboards. It allows you to create your own pages essentially and see EM data as you wish. Now, we also provide you really as samples a number of out-of-the-box dashboards. For example, we have this Enterprise Summary Dashboard. You could use it as is if it meets your needs, or, you know, you can make a copy of this dashboard and make any customizations you want to make to it. So that really is how simple it is in this new release how you create dashboards and really use the power of visualization to see the data as you want to see and to see the types of data you want to see. So hopefully this is also something that you will find very useful in this new release.
So last but not least, we have done a lot of work to improve the resiliency of EM platform in 13.5 as well. First is the near-zero downtime feature. With the monthly release update model from the EM 13.4 version, staying up with latest RU can be challenging because every time you apply an RU, you have to take a downtime. So in 13.5, we are introducing near-zero downtime patching for applying RUs on EM while it is up and running, thus making sure that the downtime you have is as little as possible. We expect that you will see up to 90% reduction in EM downtime when applying these RUs going forward. We are also making EM patching more simplified and agile. In this release, we have consolidated the distributed agent patches into a single patch for an agent. The agent system patch typically includes the agent platform plus the agent plug-in patches in one bundle. This means that overhead of applying multiple agent patches, you know, agent platform, plug-in, etc., will be reduced and save you time. The next thing is the job system. In 13.5, we will introduce a supercharged job system for power users with additional lightweight services used to prioritize specific jobs and also to enable parallel and scalable executions. We also have Autonomous Database support as an EM repository in this release. So now you can use Autonomous Database as an EM repository, and you get all the benefits of scalability, availability, and performance that Autonomous Database provides. We've also done a number of improvements in the self-diagnostics and proactive health monitoring of EM and all the sub-components. Just to name one, for example, we have the new job diagnostics dashboard to drill down into the performance and real-time status of jobs running or blocking in the system. This includes what jobs have been retried, what job queues are blocked and why, and perhaps an agent is causing the jobs to be retried excessively that's bringing the system down. So these kind of things will really make it easy for you to make sure this Enterprise Manager's platform is resilient.
I'd like to mention that we have a number of large customers like Dell, Cerner, and CDTI that are really looking forward to this release because of these new features, as evidenced by the statements that you can see. So that was a quick review of Enterprise Manager 13.5 release. I have a blog that just went out today that covers these features in more detail, and we will have several more in-depth sessions on this release. Feel free to check them out. Thank you.