PDF Preview: Accelerate Virtual Server Performance with All-Flash Storage

Maximum virtual server performance requires storage that accelerates virtual machine operation. Storage must also address the unique needs of applications running in the virtual environment. Learn how all-flash storage solves the I/O blender effect, find out what storage features are most important, and see how All-Flash FAS can deliver a rapid return on investment.

Saluting Mike, Would you please alsoa dvice how many enterprise users are running Epic on AFF8K? Tks by Henry PAN

Similar Messages

  • All-Flash Storage: Accelerated Performance, Yes, but Is It Enterprise-Grade?

    August 2015
    Explore
    All-Flash Vendors Should Meet These Criteria
    Here's our checklist to help you evaluate the enterprise-grade capabilities of all-flash storage solutions you may be considering. It helps you assess what you should be most concerned about as you invest in solid-state storage arrays for critical application environments.
    Your enterprise should settle for no less.
    AFF Enterprise-Grade Checklist
    Performance
    Audited performance benchmarks
    Quality of service
    Management Simplicity
    Single pane-of-glass management
    Seamless scale-out and scale-up
    NAS and SAN
    Data Protection / Security
    Synchronous / asynchronous replication
    Secure multi-tenancy
    F2D2C integrated data protection
    Application Integration
    Microsoft, Oracle, SAP, VMware, Citrix
    Veeam, Commvault, Veritas, TSM
    OpenStack, CloudStack integration
    When you've gone through the list, contact your NetApp sales rep or reseller to see if you qualify for our free All Flash FAS try-before-you-buy offer—it's a risk-free way to get a taste of what true enterprise-grade, all-flash storage has to offer.
    Mike McNamara
    Senior Manager, Product Marketing
    NetApp
    Sure, high performance and low latency matter. Business competition—relentless 24/7/365—demands that you coax maximum speed and responsiveness from key business operations. Faster time to market translates to greater customer satisfaction and competitive advantage.
    That's why your forward-looking IT peers increasingly embrace all-flash storage: By speeding performance and reducing latency, flash accelerates business value—driving revenue, enhancing the customer experience, and reducing costs through consolidation.
    But when it comes to all-flash storage solutions today, high performance and low latency are not enough. Your enterprise needs more.
    Turn on a Dime—Without Compromise
    In an enterprise context, "agility" refers to how well and rapidly a company can adapt to market and environmental changes—and do it in productive, cost-effective ways. Robust data management and enterprise-grade capabilities are key to staying on top of today's dynamic marketplace.
    Accordingly, astute IT bosses are asking questions around all-flash storage solutions: How available is my data? How much flexibility do I have to move that data over its lifecycle? What about data protection? Does the solution offer native support for cloud?
    In essence, they're saying, "Okay, Mr. Storage Vendor, you can accelerate my performance and reduce my latency. But can you wrap that in the enterprise-grade package that I need to run my business?"
    NetApp® All Flash FAS (AFF) answers all these questions. It's an enterprise-grade storage solution.
    Figure 1) Powered by clustered Data ONTAP, All Flash FAS is built for virtualized, shared environments requiring high performance plus robust data management.
    Source: NetApp, 2015
    Powered by Clustered Data ONTAP
    Driving All Flash FAS is NetApp clustered Data ONTAP®, the #1 storage operating system (OS).* It's the foundation of the Data Fabric, our vision for the future of data management. We architected Data ONTAP to have enterprise-grade, unified, scale-out capabilities that deliver nondisruptive operations, storage and operational efficiency, and scalability over the lifetime of the system. Data ONTAP is the basis for virtualized shared storage infrastructures.
    Backed by Data ONTAP FlashEssentials, AFF simplifies flash deployment, eliminates flash silos, accelerates database performance with 20x faster response times, and reduces your total cost of ownership. Enhanced native capabilities of the OS bring integrated inline efficiencies to AFF. These include newly enhanced compression, inline zero-block and always-on deduplication for 5x–10x average space savings.
    All Flash FAS employs the Data ONTAP log-structured file system WAFL® (Write Anywhere File Layout). It's optimized for flash media to minimize latency, reduce wear, and maximize usable capacity. The solution's optimized write, read, and parallel processing results in consistent submillisecond latency and high performance.
    Figure 2) Flash-optimized write architecture reduces latency and increases the longevity of SSDs.
    Source: NetApp, 2015
    Flash-to-Disk Agility
    All Flash FAS supports live workload migration between flash and hard disk drive (HDD) tiers—on premises or in the cloud. As data characteristics and application environments change, data that's no longer performance-driven doesn't need to be in flash. All Flash FAS gives you the flexibility to move that data nondisruptively within the same cluster to other nodes that are just spinning media for cost-effective backup or archival for example—quickly and easily.
    Flash-to-Cloud Data Protection
    NetApp is the first all-flash system to support integrated snapshots and cloning to a software instance running on Amazon Web Services. You can move your data securely across your choice of clouds—enabled by Cloud ONTAP™ and NetApp Private Storage for Cloud.
    Enterprise-grade Data Protection Delivers Resiliency
    Industry-leading, advanced data protection technologies in AFF improve your resiliency. Core features include RAID DP, in-core checksums, lost write protection, read-after write verification, and media scans that verify that data written to the solid state drive (SSD) is written correctly and regenerated.
    For enterprise-grade data availability, NetApp MetroCluster™ is the only high-availability and disaster recovery software integrated into an all-flash array. With MetroCluster, you recover from failures with zero data loss, maintaining continuous data availability for mission-critical applications—at half the cost and complexity of competitive solutions. All Flash FAS with MetroCluster is one of the only all-flash arrays that supports integrated synchronous mirroring with a single management view. And, according to Forrester Research estimates, organizations can achieve up to 143% return on investment (ROI) through MetroCluster.
    NetApp SnapMirror® software replicates to any type of FAS system—all flash, hybrid, or HDD, on premises or in the cloud—reducing overall system costs.
    Enhancements that improve SSD durability and life include metadata and user data that are coalesced in memory before being destaged intelligently to the SSD subsystem. This ability minimizes wear by reducing the amount of writes dispatched to the SSDs. Data is never overwritten in place and writes are striped across multiple SSDs, resulting in more even wear.
    Furthermore, where other storage vendors compete, we partner. One road leads to vendor lock-in; the other leads to interoperability. Through strong data protection partnerships, All Flash FAS delivers integrated support for Veeam, Commvault, Veritas, and Tivoli Storage Manager (TSM) solutions.
    Leading Application Integration Covers All the Bases
    Deep application integration sets All Flash FAS apart from other all-flash solutions on the market today.
    Delivering tight assimilation with Oracle, Microsoft SQL Server, SAP, Citrix, VMware, and OpenStack and CloudStack environments, AFF streamlines your management and data protection chores—saving you time and cycles. Application-consistent backup and recovery is integrated and automated.
    Storage Management Made Easy
    All Flash FAS includes the NetApp OnCommand® family of management software for automated tools that further simplify management of storage operations. Setting up and configuring AFF typically takes less than 15 minutes leveraging preconfigured systems for SAN deployments. OnCommand Workflow Automation automates common storage tasks such as provisioning and data protection for environments such as Microsoft SQL Server and Oracle.
    To optimize storage for peak performance and to keep everything running smoothly, OnCommand Performance Manager automatically monitors and analyzes performance. It's an integrated component of OnCommand Unified Manager.
    With multiprotocol support—Fibre Channel (FC), FCoE, iSCSI, NFS, pNFS, CIFS/SMB—AFF delivers a common look and feel across your entire heterogeneous enterprise. Network-attached storage (NAS) is especially useful for VDI workloads typical in Citrix XenDesktop and VMware Horizon View installations.
    The Need for Speed
    All Flash FAS accelerated performance enables server consolidation and can reduce database license costs by 50%.
    In a recent SPC-1 benchmark, the All Flash FAS8080EX delivered 685,000 SPC-1 IOPS, ranking in the top 5. For database environments, AFF delivers 309K IOPS at 1ms latency for Oracle database 12c and 280K IOPS at 1ms latency for Microsoft SQL Server. VDI performance is equally impressive with 160K IOPS at ~1ms latency (80% write, 20% read).
    Like every FAS controller, AFF supports all the same protocols and features as FAS hybrid arrays. This includes the ability to scale up and out to 8 nodes for FC, or 24 nodes for NAS. With multi-tenancy built in and quality of service to protect important workloads, All Flash FAS is more than capable of meeting the demands of enterprise IT—scaling up to 384TB all-flash capacity per high availability (HA) pair or up to 5PB all-flash capacity in a cluster.
    Free Try Before You Buy
    If you're evaluating all-flash storage arrays for your enterprise, you'll want to consider the questions in the All Flash Array Checklist for Enterprise Buyers (see sidebar, "All-Flash Vendors Should Meet These Criteria"). Compare NetApp AFF against all-flash storage solutions other vendors are offering. If their solutions don't meet these criteria, you will end up setting up a storage silo with all of the management challenges involved.
    NetApp is currently offering a free All Flash FAS trial for qualified enterprises. It's a risk-free way to put all-flash storage performance with enterprise-grade capabilities through its paces in your own environment. Contact your NetApp sales rep or reseller to find out more
    The NetApp AFF8000 product line delivers on all the elements in the checklist. Because yes, high performance and low latency matter—but for enterprise IT mavens, that's just the beginning.
    *IDC Worldwide Quarterly Disk Storage Systems Tracker 2015 Q1, June 2015 (Open Networked Disk Storage Systems revenue)
    Mike McNamara, senior manager of product marketing at NetApp, has 25 years of storage and data management marketing experience. Prior to NetApp, Mike worked at Adaptec, EMC, and HP. He helped lead the launch of the industry's first unified scale-out storage system, iSCSI and SAS storage system, and Fibre Channel storage system. Mike is a former marketing chairperson for the Fibre Channel Industry Association, and current member of the Ethernet Technology Summit Conference Advisory Board and the Ethernet Alliance. He frequently contributes to industry journals and speaks at industry events.
    Quick Links
    Tech OnTap Community
    Archive
    PDF

    Saluting Mike, Would you please alsoa dvice how many enterprise users are running Epic on AFF8K? Tks by Henry PAN

  • PDF Preview: All Flash Storage: Accelerated Performance, Yes, but Is It Enterprise-Grade?

    All Flash FAS simplifies flash deployment, eliminates flash silos, accelerates database performance with 20x faster response times, and reduces your TCO.

    Saluting Mike, Would you please alsoa dvice how many enterprise users are running Epic on AFF8K? Tks by Henry PAN

  • Outlook 2007 PDF preview on Windows Server 2003?

    Simple question. Does the new (Reader 8.1) PDF Preview feature for Outlook 2007 work in Windows Server 2003? Or is it only for XP and Vista? The patch notes strongly suggest the latter, but in other places (such as a blog on this site) there are notes explicitly stating that the feature should work in server operating systems.
    We happen to be running Citrix (PS4.5) on this server as well, but I'm not convinced that's material: the feature doesn't work running Outlook on the server's own console.
    When I attempt to preview a PDF in Outlook 2007, I am told, "This file cannot be previewed because of an error with the following previewer: Adobe PDF Preview Handler".
    Before I go much farther with troubleshooting this, can anyone confirm firsthand that PDF previewing works in Outlook 2007 on Server 2003?

    I was using Adobe Acrobat Professional 8.1.2 and Outlook 2007 on Server 2003 SP2 and the Adobe PDF Preview Handler worked.
    Unfortunately, it stopped working after some time and I'm pretty sure I did not make any changes to the system. However, I did try to uninstall, reinstall and update Adobe Acrobat but still, it does not work.
    Any ideas why this is happening how to resolve it?
    Many thanks.

  • Installing second Exchange server 2010 with all roles (CAS, HT and MB) in the same domain?!

    Dear all,
    first of all thanks for reading this topic :-)
    In our enviroment, we have a Exchange 2010 server (Version: 14.01.0438.000), whitch is installed with all server roles (CAS, HT and MB) on one server(OS is Windows Server 2008 R2 Enterprise). This exchange server sends all external mails to a smarthost (Redddox).
    We are using Outlook Anywhere, Active Sync and OWA.
    Now, we need to migrate this Exchange server to another one, because we think, that the server´s OS is corrupt and also, there are wrong licenses installed. The "new" server will have the same OS Version and Exchange Version (2010).
    Currently I´m a little afraid, to install a new one - because I think when I will install the CAS and HT role, something will happen in my productive enviroment (Autodiscover, SMTP Connectors, Cerficates a.s.o.).
    To install the mailbox role, I think this will not affect anything.
    Can you help me a little bit in what to take care of? Do I need to preconfigure something, before I will install the second exchange? What about the version / service pack of Exchange to install? Must it be the same as installed on the first one?
    Any help would be appreciated!
    Jennie

    Hi Jennie,
    Below are the steps if you are not planning to upgrade.
    1) Install new Exchange2010-SP3 Server all roles Please check
    this
    2) Install the certificate in the new server by requesting a duplicate from the 3rdparty CA. Or export from the existing Exchange 2010 and import to the new one. Please check
    this
    3) Set your autodoscover, OAB, ECP, OWA URLs same as the current Exchange. Below artciles will help you to do that.
    For OAB, Autodiscover, EWS please check steps 5,6,7in this
    For setting OWA and ECP URLs please check
    this.
    4) Move few mailboxes as a test and check. If no errors move the rest.
    5) Move your OAB generation server to the new server. Please check
    this
    6) Move you public folder contents to the new server you have. Please check
    this
    7) Configure your firerwall to receive emails on the new server and other services like EWS, OWA, ActiveSync.
    8)
    Add the new server as the source server and in the current send connector and remove the old server from the send connector.
    Shutdown the server for a couple of days and monitor. So you will know if you missed something.
    Uninstall Exchange2010 from the add remove programs.
    Thanks, MAS
    Please mark as helpful if you find my comment helpful or as an answer if it does answer your question. That will encourage me - and others - to take time out to help you.

  • Firefox has trouble with all flash applications

    My favorite forum website recently dropped support for IE. This specific reason, is the ONLY reason I stopped using IE. Firefox has a much closer resemblance to IE than chrome, and this is the ONLY reason I chose it over chrome.
    Everything works fine, except every and all flash applications.
    Youtube, animated ads, games... everything causes huge cpu load spikes and lags like crazy (I'm sure it would cause a crash on a basic computer). I have a very high end rig, the problem is not my hardware, or internet speed (100mbs). It is entirely fire fox.
    All flash apps work FINE on IE/chrome, I tested them. Please offer support, thank you.

    Crashes or other problems with certain multimedia content in Firefox (such as Youtube videos and Flash animations or games) can often be resolved by performing the steps in these Knowledge Base articles:
    * [[Flash Plugin - Keep it up to date and troubleshoot problems]]
    * [[Flash 11.3 crashes]]
    * [[Flash 11.3 doesn't load video in Firefox]]
    On Windows Vista and above, you can disable Flash protected mode by following the instructions on these pages:
    * http://forums.adobe.com/thread/1018071#TemporaryWorkaround
    * http://kb.mozillazine.org/Flash#Disabling_Protected_Mode_in_Flash_11.3
    (See [http://blogs.adobe.com/asset/2012/06/inside-flash-player-protected-mode-for-firefox.html this Adobe blog post] for technical details about Flash protected mode.)
    For recent updates for Adobe Shockwave Flash, please update your version to 14.
    Please refer to the security updates [http://helpx.adobe.com/security/products/flash-player/apsb14-18.html and Adobe Security Bulletin] for more information.
    <!--Please also note that html5 may need to be enabled to view some videos on youtube [https://www.youtube.com/html5]-->
    <!--[version 13]Please refer to these security issues [http://helpx.adobe.com/security/products/flash-player/apsb14-13.html and Adoble Blog post] for more information on the security update. -->
    Please tell us if this helped!

  • Unacceptable performance with USB3 flash drive on USB2 interface

    Just purchased a 64G USB3 flash drive to be used with my old nc8230 with XP Pro. The notebook's USB2 interface works perfectly with all types of USB2 devices. However, writing and reading data on the USB 3 thumb drive is extremely slow. Loading it with 42GB of data (10000 files) took me 5.5 hours!!! Tests showed, that when reading or writing a file I see an initial delay of about 2 minutes until the transfer of data starts and seems to work perfectly.
    I just updated the USB driver to the latest version available - no improvement.
    Needless to say that the problem does not show up on other PCs - showing that the thumb drive is OK.
    Handling the above mentioned 42GB data on my external 2.5" drive via the same USB port works perfectly as well (no delays).
    Any idea? Thanks

    Please cancel this question.
    In case someone were interested here some more details:
    In the meantime I made more tests and found out that the problem is caused by the new USB3 thumb drive and not the notebook. The initial good results on other PCs were caused by the point that I did not fully understand how to reproduce the failure. After this is now clear (primarily high latency times especially when generating directory entries but also when reading the directory structure), the problem showed up equally on another notebook with XP and a desktop with Win7.
    I'm now trying to get a replacement.
    Thanks, Dieter

  • Figured out the issue with local flash storage

    OK , a lot of people have had the same issue i had with the local flash storage settings not working, or any of the settings for that matter.....i tried everything and nothing would work. After hours and hours of troubleshooting i finally switched my default web browser from internet explorer to google chrome and bam all the settings were back to working order. Hope this post helps.

    bug is closed by Adobe as "NotABug".
    Flash Players before 11.3.300.257 had different behavior that was more reliable: placing "Dictionary" object to the local storage allowed reading it as "Dictionary". Like placing "Array" resulted to reading "Array" from shared object.
    With Player 11.3 it doesn't work anymore. Reading a property that was placed as "Dictionary" results to plain "Object" read. Any code like "var a:Dictionary = sharedObj.myDictionary as Dictionary;" returns "null". This is very annoying bug that resulted to our application not working for 2 days until we found the reason. I regret Adobe doesn't admit this a bug.
    Writing this in order if someone of developers face with the same issue.

  • Server Performance with Form 6i server

    Hi,
    I am still on 6i. Recently I begin to move from the traditional client-server to the web and discovered that the 6i could not operate well in Windows XP on P4 machines. So I started to put in patch 10. It works fine.
    However, I could not find a solution to stop the end users from closing their browser while the 6i applet is still running. This is casuing a major problem to us becasue when ever a user does this, my server's processor indicator will shows that my processor is running at 100% and the memory (both virtual and physical) will increase unlimitedly until cilling and the entire server will jam and services will stop.
    Then when I tried to restart the services, the server will report that the port is being use and cannot start the service. Result of so, I will have to wait for 30-40 minutes and all my users are interrupted.
    I have even tried to tell the users to exit the application prior to closing the browsers, but some of them were the directors and they just being too buzy to follow the steps and would rather choose to close the browser and continue with their other routines.
    Can any one share some ideas to protect this? Thanks
    Thank you

    I shall try that, but jon, for your information, I cannot patch 15 becasue with patch 15, I will be facing the same problem with P4 on xp again.

  • Poor Bluetooth performance with all aluminum Mac Pro

    The all aluminum case for pre-mavericks Mac Pro is the primary reason for poor antenna performance of Bluetooth. Aluminum, being a good conductor, acts as a reflector of EMF forming what is known as a Faraday Cage, essentially trapping emissions within the box. Normally, holes drilled in the case would allow enough EMF to pass through the case but the hole size for transmission is directly proportional to the lowest frequency that will pass through easily. An example is the typical microwave oven where the window is actually a conducting sheet with small enough hole such that higher freq light waves pass easily through the sheet but lower frequency microwaves do not. Perhaps if the drive bay covers were made of plastic, transmission of the Bluetooth and wifi would be sufficient. One might carefully nibble a slot or two in the side door but this would surely invalidate any warranty and care would have to be taken not to get aluminum particles inside the computer. This would probably nullify any FCC approval of the machine due to clocking noise interference with other devices. The best approach would probably be a tuned waveguide that passes frequencies only around 2.4 GHz ~ 5 GHz through the case (very similar to the way tuned ports on audio speakers pass a band of frequencies needed to extend bass response). I'm sure Apple will have corrected this design flaw in the new Mac Pro introduced this year. If you have lots of time on your hands you can get a college level text on electromagnetics (Maxwell's Equations, etc.) and derive all the details but with a bit of experimenting and reading about tin can antenna mods to wireless routers you should be able to improve Bluetooth performance significantly.

    Thanks for the suggestion.  I will have to move my wireless home base station.  It is on the same desk as my mini so much less than 5 feet away. 

  • Sun Java Application Server Performance with Web Services

    We are running a web service on SJAS (Standard Edition). Load testing under 1 or 2 users works alright, but with 5 users making concurrent web service calls it immediately crashes the domain. Has anyone else run into performance issues when using SJAS with Web Services? Are there any configuration parameters we need to setup to handle more than 1 connection?
    I'm sure that the application server can handle many concurrent connections, just not sure how to configure and where to look for more information. If anyone is aware of actual numbers for load testing, that would be great information.
    Thanks,
    Dawson

    Hi Dawson,
    SJSAS can definetly handle more users. Can you please tell us, what version of SJSAS and web services (jax-rpc 1.1 or jax-ws 2.0?) implementation you are using? When you say it crashes the domain, do you see any error messages in the server.log?

  • PDF Preview for a  PO template with polish characters shows question marks

    When I view the PDF using application I see the polish characters(Boilerplate text) as question marks. But when I view it through XML Publisher Dsktop version 5.5.0 it is fine I have download the new version of adobe 7.08 and I can see the polish characters but the preview in XML publisher in the application stll shows "?"

    No I havent configured any fonts on desktop/application .. when I view Document properties on the PDF which works the XML Publisher version 5.5.0 and that of the application is 5.6.0.
    In the fonts tab on the desktop version there is one additional one listed fromt he application one - AlbanyWTJ(Embedded Subset)

  • Server performance with beans in WEB-INF/classes

    Hi,
    Someone over at oracle corp told me that if i put my beans in WEB-INF/classes, my server would notice a significant performance degradation since the beans would be automatically refreshed or dynamically loaded everytime.
    WHat I am doing now is restarting the webserver everytime a bean is created or changed.
    My goal was to have the new beans automatically noticed by the web server so i didnt have to stop and restart.
    Anyone have any input for me.
    Thanks

    Thanks I will do that but just one question.
    What about when the site it live. Say the boss comes to me and says I need this changed.
    I change the bean, compile it, and MUST restart the web server even though there could be many people on the site. Is that what everyone is doing?
    I think I am a little confused because in a production system I dont feel that the web server should ever be stopped unless there is some type of failure.

  • Sending pdf as an attachment in email with all changes done on the pdf

    Hello Everyone,
    I have developed a webdynpro java application which sends a interactive form as an attachment in an email. Everything works fine except that there are some editable fields on the form and the changes dont go in the attachment. The Send Email button is on the webdynpro View. I tried many different things but nothing worked. Somehow the changes done on the form dont get saved in the pdfSource when a button on the webdynpro view is clicked. Any help would be very much appreciated.
    Thanks,
    Saika.

    Hi Hari,
    check this link: http://help.sap.com/saphelp_nw70/helpdata/en/51/d5cd16235e4643ae8ec92395c4ad97/frameset.htm
    Also search on SDN, you will get many thread links with same question
    ABAP Proxy Attachment
    Regards,
    Praveen Gujjeti

  • Disappointing query performance with object-relational storag

    Hello,
    after some frustrating days trying to improve query performance on an xmltype table I'm on my wits' end. I have tried all possible combinations of indexes, added scopes, tried out of line and inline storage, removed recursive type definition from the schema, tried the examples from the form thread Setting Attribute SQLInline to false for Out-of-Line Storage (having the same problems) and still have no clue. I have prepared a stripped down example of my schema which shows the same problems as the real one. I'm using 10.2.0.4.0:
    SQL> select * from v$version;
    Oracle Database 10g Enterprise Edition Release 10.2.0.4.0 - Prod
    PL/SQL Release 10.2.0.4.0 - Production
    CORE 10.2.0.4.0 Production
    TNS for Linux: Version 10.2.0.4.0 - Production
    NLSRTL Version 10.2.0.4.0 - Production
    You can find the script on http://www.grmblfrz.de/xmldbproblem.sql (I tried including it here but got an internal server error) The results are on http://www.grmblfrz.de/xmldbtest.lst . I have no idea how to improve the performance (and if even with this simple schema query rewrite does not work, how can oracle xmldb be feasible for more complex structures?). I must have made a mistake somewhere, hopefully someone can spot it.
    Thanks in advance.
    --Swen
    Edited by: user636644 on Aug 30, 2008 3:55 PM
    Edited by: user636644 on Aug 30, 2008 4:12 PM

    Marc,
    thanks, I did not know that it is possible to use "varray store as table" for the reference tables. I have tried your example. I can create the nested table, the scope and the indexes, but I get a different result - full table scan on t_element. With the original table I get an index scan. On the original table there is a trigger (t_element$xd) which is missing on the new table. I have tried the same with an xmltype table (drop table t_element; create table t_element of xmltype ...) with the same result. My script ... is on [google groups|http://groups.google.com/group/oracle-xmldb-temporary-group/browse_thread/thread/f30c3cf0f3dbcafc] (internal server error while trying to include it here). Here is the plan of the query
    select rt.object_value
    from t_element rt
    where existsnode(rt.object_value,'/mddbelement/group[attribute[@name="an27"]="99"]') = 1;
    Execution Plan
    Plan hash value: 4104484998
    | Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
    | 0 | SELECT STATEMENT | | 1 | 40 | 2505 (1)| 00:00:38 |
    | 1 | TABLE ACCESS BY INDEX ROWID | NT_GROUP | 1 | 20 | 3 (0)| 00:00:01 |
    |* 2 | INDEX RANGE SCAN | SYS_C0082879 | 1 | | 2 (0)| 00:00:01 |
    |* 3 | FILTER | | | | | |
    | 4 | TABLE ACCESS FULL | T_ELEMENT | 1000 | 40000 | 4 (0)| 00:00:01 |
    | 5 | NESTED LOOPS SEMI | | 1 | 88 | 5 (0)| 00:00:01 |
    | 6 | NESTED LOOPS | | 1 | 59 | 4 (0)| 00:00:01 |
    | 7 | TABLE ACCESS BY INDEX ROWID| NT_GROUP | 1 | 20 | 3 (0)| 00:00:01 |
    |* 8 | INDEX RANGE SCAN | SYS_C0082879 | 1 | | 2 (0)| 00:00:01 |
    |* 9 | TABLE ACCESS BY INDEX ROWID| T_GROUP | 1 | 39 | 1 (0)| 00:00:01 |
    |* 10 | INDEX UNIQUE SCAN | SYS_C0082878 | 1 | | 0 (0)| 00:00:01 |
    |* 11 | INDEX RANGE SCAN | SYS_IOT_TOP_184789 | 1 | 29 | 1 (0)| 00:00:01 |
    Predicate Information (identified by operation id):
    2 - access("NESTED_TABLE_ID"=:B1)
    3 - filter( EXISTS (SELECT /*+ ???)
    8 - access("NESTED_TABLE_ID"=:B1)
    9 - filter("T_GROUP"."SYS_NC0001300014$" IS NOT NULL AND
    SYS_CHECKACL("ACLOID","OWNERID",xmltype('<privilege
    xmlns="http://xmlns.oracle.com/xdb/acl.xsd" xmlns:xsi="http://www.w3.org/2001/XMLSchema-insta
    nce" xsi:schemaLocation="http://xmlns.oracle.com/xdb/acl.xsd
    http://xmlns.oracle.com/xdb/acl.xsd DAV:http://xmlns.oracle.com/xdb/dav.xsd"><read-properties
    /><read-contents/></privilege>'))=1)
    10 - access("SYS_ALIAS_3"."COLUMN_VALUE"="T_GROUP"."SYS_NC_OID$")
    11 - access("NESTED_TABLE_ID"="T_GROUP"."SYS_NC0001300014$")
    filter("SYS_XDBBODY$"='99' AND "NAME"='an27')
    Edited by: user636644 on Sep 1, 2008 9:56 PM

Maybe you are looking for