Oracle Enterprise Manager Technology Forum 2023: latest innovations and product roadmap

Oracle Enterprise Manager Technology Forum 2023: latest innovations and product roadmap

Show Video

thank you hello everyone thanks for joining us today for this Oracle Enterprise Manager technology Forum 2023 start we have some great content planned for you today I am joined with mughese minhas who is the senior vice president of product management for Oracle Enterprise Manager we started doing this forum last year so this is the second edition of this we had some great viewership last year some great content but I think this year will be even more exciting so for the next three days we start with the keynote right now and give a little bit of a sneak peek or a sneak preview of what's coming next year we have three customers this week we have atnt we have Onyx Enterprise Solutions and Dell and we love providing customer content to to you the customers because having them speak for our products always gives a bit more of an extra impact and nice to get a third-party validation of the work we've been doing and then there will be a few General Sessions in the next few days and in total about 20 technical sessions so we hope you will enjoy it this year just like last year so a little bit of recap Oracle Enterprise Manager what are sort of the core capabilities that we've been working on for over two decades first off one important aspect is hyperscale Right a lot of our customers are of course the big Enterprise customers small medium business as well but certainly the big Enterprise customer base lots of Oracle databases Oracle middleware and so forth and so we have customers that literally have millions of Targets target being an operating system the database a middleware stack and so we can manage a complete Enterprise with one product one single pane of glass that can monitor and manage all of these different targets so we scale very well and one of the advantages we have that we don't often talk about is that Oracle is one of the top three users of Oracle Enterprise Manager so we basically use our own product to manage our own I.T infrastructure and that allows us to deploy early versions of the Next Generation and give internal feedback before we hand it over to customers and and let the customers work with it so we have some great internal usage as we do with pretty much all the products we work on across the company secondly a completely integrated full stack management and what we mean by that is we we really can see end-to-end into the Oracle stack we can do things like patching upgrades installations monitoring management performance viewing everything every aspect of what you would need to do with our products you can do at Oracle Enterprise Manager so really full stack support and of course engineered systems Oracle exodata the CDL array Appliance is completely managed end-to-end you can see the firmware versions you can see the host the storage nodes everything so one pane of glass has always been a very important part of what we do with em and then of course lastly comprehensive database management we we built the Oracle database so we also manage it we have the same developers in some cases writing code that is in the Oracle database itself and also write code on Enterprise management it's really tight integration and that gives us the best visibility of the inner workings of the database and allows us to do things that no other database management software company can do compared to us so the last 12 months what have we done well obviously invested a lot in em and that shows by having 10 release updates or patch updates done over the last year on 13.5 over 400 450 460 new features introduced during those 10 Rus and over 400 bugs fixed and what's interesting here is that you know we started out with Oracle Enterprise Manager 13.5 we started out in a new model a new development model where it's more the cloud delivery model regular updates but regular updates not just with bug fixes but also new features so when the developers finish something it gets integrated and gets released gives us a few advantages number one the amount of testing that we have to do on each given release is focused on just the code that has changed secondly it means that we can show customers for one we can show you that we continue to invest in this product and it gets enhancements and new functionality and you don't have to wait 18 months and or 12 months or 18 months you get them on a regular basis and also by slowly introducing new features there's no one big new big dump of software that takes a lot of time to in to digest and Implement within the company so those different pieces have really helped us have helped us with stability and also have helped us with customers that feel like okay there's a lot of cool stuff happening and they feel very comfortable with Oracle Enterprise Manager there's a lot of stuff coming again in the next 12 months and mugees will will spend some more time going into into detail of some of those things it's uh it's it's a it's a good feature set and what I really quickly want to talk about is Enterprise Manager 24c but first off a a one of the changes we recently made is that Enterprise Manager 13.5 support is going to be extended all the way through 2026. and one of the reasons we're doing that is because with em24c coming a big version change a new major release we want to give our existing customers the time to stay with 13.5

learn em24c changes and then as time progresses have have a little bit more chance to test before doing the upgrade so 24c what are the main things we're doing here first of all we're working on enhancing the architecture of Enterprise Manager and what I mean by that is we're we're going you know we already started the development model of regular updates sort of the cloud model we're doing the same thing from an architecture point of view so we have container images for specific pieces of Enterprise Manager we're not going crazy with microservices where we have hundreds of little microservices for containers we'll just have a handful three four five but the the big functionality pieces will go into a container one of the advantages with that model is that we will be able to do upgrades with zero downtime and a a customer complaint if I can can use that term that we've had over the years is that Enterprise Manager is used to provide alerting and monitoring for systems so if the customer has to upgrade Enterprise Manager itself and then incur a downtime for that let's say it takes half an hour then events coming in from production systems are not caught because the monitoring software is down so a major change with 24c will be that we separate out the alerting and job system into a into a set of containers and so as em itself the core gets upgraded these will keep running and you will not lose any alerts or any incoming messages from from an important production application that you were you were monitoring so that's a I would say one of the biggest advantages that will come with 24c then secondly we we will continue doing the regular release update model and also with 13.5 after 24c is released we will continue doing new updates on 13.5 for a good period of time as well so it's not it's not going to just stop next year with 24c just to be clear and anyway with 24 we're going to continue the same development model and we're also going to do more around jet Oracle jet is the is our modern toolkit that we use for Fusion applications and and most of the the cloud interfaces that we have for for application performance monitoring and operations insights and so forth and so with with jet being a modern UI very nice widget user-friendly widgets what we are able to do is when we develop functionality for database Management in our Cloud management component we can take that code and also deploy it in Enterprise Manager a certain subset of those and the opposite when we are doing new function development and Enterprise managed for 24c then we can likely take that code and also deploy it in Cloud so we're able to address two Targets with one with one framework and so it lightens the development workload and it allows us to basically address two different deployments with one effort then simplified lifecycle management this is pretty obvious with container management we we basically provide a new container image and you can update that image so you don't have to go and go to the full install cycle of of the product and of course we always focus on improving security wherever wherever we can so that will continue obviously as well in terms of functionality I already talked a little bit about zero downtime so we we will make sure that alerting and ticketing will will continue even while we're doing upgrade mentioned Oracle jet so it'll be a very modern very responsive UI you can do the back button forward all that stuff will work really really well and seamlessly Cloud extensions we started down this road um down this path a few years ago now and so Enterprise Manager is doing more towards Oracle Cloud whether it's managing targets that are living in oci or whether it's new functionality where we Bridge the two so we we take data from Enterprise Manager we send it to the cloud service and you can then we can then do more complex things like Warehouse operations and exadata insights and so forth on the cloud side so we we basically provide two parts to the management solution Fleet Automation and compliance Fleet automation has has always been been very important as we get larger and larger customers certainly with exadata we we have to be able to manage the large ecosystems of Hardware software the full stack ml is everywhere with everything we do whether it's in cloud or on premises we bring machine learning into our products and so with Enterprise Manager 24c you will see a lot more machine learning coming into play as well and lastly we have new connectivity models coming for management and we introduce agent SQL routing and the jdbc HTTP proxy some of the advantages here is that we will make it a lot easier to connect Enterprise Manager from one side to another data center or into Oracle cloud or other Cloud vendors as well so much simpler Network configurations required for for Enterprise Manager to talk to all the different targets now with all that talk about 24c we go back to 13.5 and for that Magee's will give you a good overview of what we have coming over the next 12 months thank you very much so thank you m so I'm going to give you a quick overview of the new features we have added in em-135 in the last year most of the new features we have added can be grouped in three main themes they are hybrid Cloud management Ops Automation and extensibility and resilience of the core em platform let's start with hybrid Cloud management so the first feature I will talk about is Adam Spotlight as I'm sure you know the the basic atom feature provides expert level DB time-based performance analysis on a pair of AWS snapshots and this feature has been there since Oracle database 10g because atom analysis is typically done on AWS snapshots immediately after a snapshot is captured the findings and recommendations tend to be tactical in nature as a normally only cover a single hour so what Atom Spotlight does it is that it Aggregates Adam findings and recommendations over a longer period of time could be days or weeks and helps identify or put a spotlight on things like systemic problems recurring bottlenecks anomalous behaviors and as a result provides a more strategic perspective on your workloads so when you want to see how a database has been performing over the last day or a week the right place to look at is atom Spotlight it will show you what eventful things have happened what kind of things require further analysis I will show you a short demo of atom Spotlight in just a little bit the next feature we have made a number of enhancements is in the area of real-time SQL monitoring this is a broadly used feature as you well know we now not only show you the full execution plan but also highlight problematic or expensive plan lines in the execution plan so you know where to focus your training effort we've also added an Optimizer environment section which displays the parameters used to generate the execution plan and we also have an outline tab that can be used to create a plan outline to reproduce the exact execution plan if you needed to uh We've also added a graphical execution plan now that allows you to view uh the execution plan in in a graphical format as opposed to a table format uh it's made up of collapsible tree of nodes with each node represents an operation uh in this plan you can display as much or as little data as needed and you can rotate the tree either vertically horizontally to see the information how you like to see it you can also expand the plan region in the full screen mode which is very useful for large execution plans next is the chargeback feature so we have made the chargeback feature oci aware so now you can divide the cost of oci entities and charge internal customers based on their respective usage this includes our chargeback support for unused resources that is resources that are idle or not being used can now be charged to the same entity or Associated entity within chargeback second we have added a chargeback support for fixed cost item type so under this plan you can include fixed costs involving operating resources such as you know power real estate rents admin salaries and so on to your chargeback model so that's with respect to chargeback Performance Hub which once again is a very broadly used feature we have added to that the exadata tab so this makes performance Diagnostics of exadata or databases running on X data much easier it provides you a unified view of exadata hard disk and flash a disk perform statistics it also provides insights into the health and performance of all the system components including databases Excel data storage cells ASM CRS and more so the next feature I want to discuss is hybrid database as a service so Enterprise Manager now supports pooling of infrastructure resources on-prem across on-prem assets and exadata Cloud which includes executed a clouded customer or the exit data cloud service for building out your private hybrid Cloud to enable the devops users to deploy databases on demand this makes making databases available for new applications development and testing easy fast and automated last feature in this section is awr Explorer this feature is yet to be released but will be out soon so awr Explorer enables dbas to perform Advanced performance analysis of awr data with awr Explorer you can slice and dice awr data in many different ways to quickly detect root cause of problems instead of generating the old HTML report which I'm sure you're already familiar with you can now interact directly with the awr data using this intuitive Explorer interface and analyze the data across multiple awr snapshots one of once again one of the limitations of awr report is that it really is analysis across two AWS snapshots but with the Explorer you can look across all the snapshots that are stored in the repository you can zoom in zoom out looking at the performance Trends if necessary you can drill down into the weight events in the histograms of the vitamins you can investigate a narrow time window and you know based on the event histogram weight spread you can understand the problem better and so on additionally if you want to generate the standard awr reports Ash report SQL report this awr Explorer interface also lets you do that and you can also save the charts that you see in this Explorer in PDF format and share it with your colleagues for you know collaborative purposes so now we'll show you a demo of Adam Spotlight which I recorded a little bit earlier optimizing performance involves change applications change workloads change data growth DBS need to understand costs and benefits of making system changes in order to boost performance so this demo shows how atom Spotlight can be used to quickly assess several different paths for improving database performance starting on the database home page that I'm sure you're very familiar with you can see the atom Spotlight link and on the recommendation style you can also access atom Spotlight from the performance menu as well I will click on the atom Spotlight link the atom Spotlight page has tabs for findings recommendations database parameters findings are displayed by default the summary timeline shows finding counts by category over the time period for each finding name the overall impact a Maximum Impact and the frequency of occurrences reported finding impacts in Adam are measure of DB time as you well know the table is sorted by overall impact as you can see here the first thing I'm going to do is I will change the time period from 24 hours to 7 days so here I go I click on last seven days now I'm looking at seven day time period so this view summarizes findings um over seven days there are 167 different atom occurrences here these aggregated findings are the key to atom Spotlight turning tactical analysis into strategic insights the chart can be overlaid with notable events by clicking on this include event no annotations checkbox small icons appear as you can see here over the timeline where instance restarts or parameter changes have occurred by hovering on this icon we share more details with you we see that this is a routine change to the resource manager plan when entering the maintenance window see the regular occurrence of paired events when this parameter is both set and unset looks like there is a four hour maintenance window being used on this database the summary timeline can be filtered by a finding category so here I'm going to click on this and I'll see the different categories by which I can filter the events the findings by and I will pick top SQL the timeline and table of findings are now filtered to only show findings by top SQL category we see that the top SQL filing happens all the time which is normal since findings are about DB time and SQL is responsible for most if not all of DB time typically there are 38 recommendations for this finding seen in the rightmost column here click on then we'll click on this number to get more details Adam makes a separate recommendation for each SQL ID included in the top SQL finding we see that tuning the very top SQL statement here results in very large overall benefit about 56 potentially half of all DB time can be saved by tuning this one SQL statement the recommendation is to run the SQL tuning advisor shown here which is integrated with atom Spotlight so we click here we run the C we can run the SQL tuning advisor and this will launch the SQL tuning advisor task I want to actually run the advisor which I'm sure you're very familiar with that we'll cancel out of here and then go and look at the second set of findings so the second finding is about commits and rollback let's drill down to see what Adam has to say here so here you can see that Adam has two recommendations indicating same potential benefit of about 13 percent um you can achieve it one of two ways either you reduce the rate and size of commits or by improving the higher performance of redo logs they have the same benefit because they are alternatives for affecting the same amount of DB time but they have vastly different implementation constraints one involves analyzing and modifying the application whereas the second one is about changing your infrastructure to improve your i o throughput atom Spotlight in this way is a great assistant in weighing the trade-offs of such widely different strategies if restructuring your app is more plausible for you take that route if you prefer the change to your ISS subsystem that's how it can also produce a similar benefit recommendations and their associate rationals have considerable detail supporting their conclusions I'll just the last one I will show is the third set of finding which is top segments by user ion cluster there are four recommendations here here we see that the top two recommendations are similar concerning i o to different partitions of the orders table since these recommendations happen almost all the time you know 162 out of 167 time they are not intermittent and benefits will accrue consistently over time uh the recommendation is to the segment advisor which helps you defragment a fragmented partition it also can help you reduce the size of the different partitions which will help you improve the i o characteristics of these particular segments and partitions so we will close this and now go back to the top of the atom Spotlight page and I'll just summarize here that you know how I've given you an idea on how atom Spotlight works and how to use it it really is designed to assess your workload Behavior over longer time periods it helps analyze performance Trends and anomalies so you can take action that improves the overall performance of your database in the area of Ops automation there are quite a few new features as well first one is top activity like this is really a simplified version of Ash analytics feature optimized for efficient real-time performance Diagnostics even under heavy database loads so it provides a simple Compact and efficient knock-like page that can be displayed on a large screen for easy visualization of your recent activity it has Auto refresh and browser settings retention capabilities in it that come in handy in observing you know real-time database performance Trend without any human interaction so the ash analytics page tends to be the data in that is more spread out is more coarse grained than fine grain top activity light has more compact data it's very similar to the way the top activity page used to be and I think a number of you asked us to you know to get that back and now based on popular demand the top activity page is back so the next feature is remote agent this is yet another feature that is not out yet but will be coming soon with the remote agent now you will be able to remotely Monitor and manage an Oracle database and all its related components like listener ASM and so on you no longer have to deploy an agent on every single database host instead you can rely on a smaller set of agents to remotely Monitor and manage your entire database Fleet this greatly reduces the overall volume of Agents you have to maintain manage and deploy we have also added a number of new exadata Cloud health and Security checks for availability and compliance so em now extends the autonomous Health framework extra check for exadata cloud which includes scanning for performance and reliability issues of all components it also provides automated and proactive notifications for any issues identified next we have introduced the new Fleet Maintenance Hub now the Fleet Maintenance Hub is designed to be your One-Stop shop for enhancing security posture of your database assets so you will get automated prioritized security patch recommendations based on the quarterly critical patch updates published by Oracle finally we have support for the for the next generation of Stig regulatory security compliance standards as well as a refresh of Industry standards for example the center for Internet Security or CIS in addition we have also integrated configuration and sensitive data assessment using dbsat for Fleet level operations so with this you can keep your databases secure and be compliant with your company policies okay so the last area and the third one is extensibility and resilience so there's much to talk about here as well we now Provide support for privileged access management or Pam integration so if the passwords for your database or host are managed in Pam providers like cyber Arc hashicorp and so on for security and compliance reasons you can simply integrate your Pam provider with em in three simple steps the Pam integration enables a seamless credential access from an external store for performing you know normal em activities like log into database from the console database patching running host commands in accordance with your company security and compliance policies I have a demo for this feature also that is coming shortly we had new rest apis for user management credential management metrics incidence Target management as you well know rest apis facilitate maximizing the efficiency and automation of your infrastructure monitoring and management within the devops pipeline so by leveraging these rest apis you can streamline your processes save time and enhance your overall productivity especially if you have already have tools like ansible or puppet or terraform in your environment you can simply call these rest apis to perform management operations now with ease integration with web Hooks and slack so with events or incidents raised in EM you can now use webhooks to share these events or incidents with external systems such as ticketing systems or you could use the slack notification method to send slack messages to your it Ops teams enabling you to explorate resolution of these incidents there's a secure and compliant process for patching plug-in deployment and installation using non-sys users so this is where you know the EM maintenance operations such as patching plug-in deployment and so on used to require assist user Navi support doing it with the nonsense user so this will provide for improved security and compliance adherence and this was yet another thing that our customers had been asking for and we've added that now the last thing I want to talk about is em diagnostic run books so this run book is is a predefined runbook is meant to help em admins you know people who manage and maintain em infrastructure OMS repository agent and so on for them to diagnose and resolve incidents on the EM infrastructure itself so steps in this runbooks include you know checking the CPU and I O of the repository database to see if it can handle the load checking for any errors in EM internal status collection jobs checking the aw report for repository database to see if the SGA PG are sized correctly and so on checking the loader threads to make sure they can handle the workload so with this run Book Em admins can quickly identify and resolve some of the more common issues seen in large em deployments finally we have now support for the cross endian em repository migration so we have a migration utility now that leverages data pump to export repository data from one platform to another with a different ndn format so this feature two has been requested for some time and you know we've finally added it to uh so with this now I will go to my second demo on the Pam integration which I also recorded a little while ago em spam integration is a great capability for securely retrieving the credentials that are stored in an external bam provider like cyber Arc or Hershey cop and so on Pam integration with em consists of three simple steps first is the BAM script registration step so here I have the Cyber script provided by my security admin and I will use that register in EM to securely retrieve the credentials so step one to register the script you have this emcli config credential provider command passed with all the parameters such as provider name input file path and so on let me execute this command so here in the output you can see that this credential provider cyber Arc is not configured in EM this completes the first step two is to create credential mapping so here we are mapping the script output to the credential type in EM let me execute the EMC like command with parameters such as mapper name credential type Etc to map the DB credential type with the script attributes so I run this and here in the output you will see that the credential mapping is saved for the database this completes the second step in bam integration with em the step 3 is the final step in the integration process to create or modify a name credential in EM here I will use a new named credential by specifying parameters such as alternate credential store credential key credential mapping to indicate that the credential for the database will be fetched from an external Pam store instead of fetching it from the EM Repository I execute the command and here you can see that it says that the sys underscore cyber R credential is now created with this the Pam integration with em is complete and you are ready to go I am going to show you how to use the spam based new name credential that we created to log into a database from encomp em console and we will see how em seamlessly fetches their credentials from a Pam provider without any extra effort so here is your familiar EMCC home page and let's say I want to run Ash analytics or run a job which will require a login so let's go to setup security name credentials and if you scroll down here you will see the sys underscore cyber R credential I have created the password management for this credential is done through cyber Arc and is not stored in Enterprise Manager now let's see how to use this credential to log into the database to view the performance page so if you go to targets let's click on databases we are going to pick this specific database and now let's say I want to go to Azure analytics so let's do that performance Hub Azure analytics so now you get to the familiar database login page here I'm going to specify the sys underscore cyber R credential that we had created earlier and now you click login and there you go directly to the page so it basically behaves exactly like an em credential would except that in this particular case the passport management is done through an external provider like cyber Arc in this case as opposed to within em itself so the behavior is exactly what you're familiar with once you have carried out those three steps I showed you a little bit earlier so thank you for joining us today foreign hey did you like what you learned make sure to check out this video or this link and of course subscribe now

2023-07-06 12:29

Show Video

Other news