Seeking advice on Zone design

I have a ZCM Zone originally built with ZCM 10.2 and has been updated over the years to 11.2.3 and soon to be updated to 11.3.1
The Zone services 15,000 devices with over 30k users (school district). The network topology is central, each school has a 10GB fiber link to the data center. Entire Zone is build on a VMWare vSphere 5 platform.
Current design consists of 1 dedicated ZCC server, 1 dedicated Inventory server, 2 dedicated image servers, and 12 Authentication/Content/Config servers. Database is SQL 2008 R2 on it's own VM. Typical guest machine config uses 2 vCPU 8GB vRAM. All run on Windows 2008 R2 SP1 SQL server uses 2 vCPU and 24GB of vRAM. DB has grown to 43GB in size (gets up to 60+GB before DB maintenance operations are run).
In ZCM 10, the closest server rules were setup to split the user traffic among four selected servers for the site. When the closest server rules allowed for groups, it was enabled to get the round robin functionality. Was never able to get the needed data from my customer to fully implement Locations, so Locations Lite is in use. Pretty much set the default closest server rule to group all 12 Auth servers in a single group. It has worked to split the load quite well among the 12 Primary servers.
Only the ZCC server and Image servers had their entire VM memory reserved (per VMWare best practice for a Java app). Was unable to reserve memory for all guest machines since it would cause to much performance issues with other guests when doing so. Because of this, I am thinking of swapping out the 12 Primary servers for 12 Satellite servers .. but I am unsure of the sanity of doing such a change. The satellite servers would run in the same virtual environment as the Primary servers.
My hope in doing this change is to improve the authentication speed, and satellite servers seem to be faster in getting the job done. Also reduce the amount of work the database server is doing by reducing the amount of Primary servers talking to it.
The change almost seems pointless, so I wanted to see what other thought about doing such a change.
thank you

We definitely want all of the VMware Memory Reserved.
Consider Converting the 2 Dedicated Imaging Servers to Satellite Servers
with the Imaging Role. This will consume far fewer resources and they
memory for Satellite Servers is not required to be fully dedicated.
12 Auth/Content/Config servers is far more than what is necessary for
15,000 Devices. Especially with 8gb of RAM. As a Test, Remove a couple
of these servers from the "Server Group" and test performance.
You may also be able to reduce the RAM from 8GB to 6GB on the remaining
10 servers to allow for dedication.
The key is that assigning RAM above and beyond what is dedicated can
lead to stability issues and will not be fully dedicated.
It is quite common for servers to fail upgrading or crash after upgrades
when the RAM is not dedicated because the servers now start hitting and
trying to use the non-dedicated RAM that was previously not used.
Also Drop an Email to [email protected]
I want to email you a utility, but will need your email address.
Note: Location Lite is just fine.
On 7/15/2014 4:56 PM, Provogeek wrote:
>
> I have a ZCM Zone originally built with ZCM 10.2 and has been updated
> over the years to 11.2.3 and soon to be updated to 11.3.1
> The Zone services 15,000 devices with over 30k users (school district).
> The network topology is central, each school has a 10GB fiber link to
> the data center. Entire Zone is build on a VMWare vSphere 5 platform.
>
> Current design consists of 1 dedicated ZCC server, 1 dedicated Inventory
> server, 2 dedicated image servers, and 12 Authentication/Content/Config
> servers. Database is SQL 2008 R2 on it's own VM. Typical guest machine
> config uses 2 vCPU 8GB vRAM. All run on Windows 2008 R2 SP1 SQL
> server uses 2 vCPU and 24GB of vRAM. DB has grown to 43GB in size (gets
> up to 60+GB before DB maintenance operations are run).
>
> In ZCM 10, the closest server rules were setup to split the user traffic
> among four selected servers for the site. When the closest server rules
> allowed for groups, it was enabled to get the round robin functionality.
> Was never able to get the needed data from my customer to fully
> implement Locations, so Locations Lite is in use. Pretty much set the
> default closest server rule to group all 12 Auth servers in a single
> group. It has worked to split the load quite well among the 12 Primary
> servers.
>
> Only the ZCC server and Image servers had their entire VM memory
> reserved (per VMWare best practice for a Java app). Was unable to
> reserve memory for all guest machines since it would cause to much
> performance issues with other guests when doing so. Because of this, I
> am thinking of swapping out the 12 Primary servers for 12 Satellite
> servers .. but I am unsure of the sanity of doing such a change. The
> satellite servers would run in the same virtual environment as the
> Primary servers.
>
> My hope in doing this change is to improve the authentication speed, and
> satellite servers seem to be faster in getting the job done. Also
> reduce the amount of work the database server is doing by reducing the
> amount of Primary servers talking to it.
>
> The change almost seems pointless, so I wanted to see what other thought
> about doing such a change.
>
> thank you
>
>
Craig Wilson - MCNE, MCSE, CCNA
Novell Technical Support Engineer
Novell does not officially monitor these forums.
Suggestions/Opinions/Statements made by me are solely my own.
These thoughts may not be shared by either Novell or any rational human.

Similar Messages

  • Seeking advice with concept / design of a Java Web Services Application

    Hi all,
    After a week of searching the internet I'm not sure how which projects, services, etc. I should use to develop my application. Please could someone offer some advice?
    Application outline (Java application running on Linux):
    1. Wait for an instruction from a Windows WCF application. Instruction contains a list of domain names and one or two other parameters.
    2. Perform queries upon those domain names (find if they are registered, etc. - takes 10-20 mins to do complete list)
    3. Send back results to WCF application.
    Solution 1 (first idea):
    1. Create SOAP web service using Java Web Services / Apache to listen for requests. Executes a Java Client Application upon request.
    2. Java Client Application performs the queries...
    3. Java Client Application sends results to the Windows web service.
    Solution 1 Problems:
    a) Using 2 applications, they won't be able to share the same memory.
    b) The whole process will be slow (having to pass the domain names from the service to application and execute the application each time) - A quick response is critical.
    c) Would like is for the whole process to be done under one application, sharing the same memory.
    Solution 2:
    1. Create a Java Daemon from scratch listening for an incoming SOAP message (no web service like Apache/Tomcat involved).
    2. Query the domain names in a new thread inside the Java Daemon
    3. Send the results back via SOAP.
    Solution 2 Problems:
    Cannot find examples of how to create a SOAP service from scratch. E.g. creating a WSDL file based on my application; converting application methods to SOAP-callable methods easily (without writing a framework).
    With all the Java tools and projects out there - Java EE, Glassfish Project - there must be a very easy way to achieve this seemingly simple task. Please can someone offer some advice?
    Many thanks for reading this.
    Richard
    Edited by: jazzinbrazil on Mar 30, 2009 4:58 AM

    You just need an app server like Tomcat.If I'm not wrong, Tomcat is a Servlet container.
    Servlets aren't deactivated when they don't receive
    any request for some time?
    How can I deploy an application to Tomcat in order to
    keep it always active?I don't know what you mean. Tomcat is an application that is always running. In what way are the Servlets deactivated?
    Apache Axis: http://ws.apache.org/axis/
    Yes I'm collecting some info about this... let's see
    it!
    Finally, to be more clearer... I don't want to start
    a new application at each invocation (something like
    getting the request, instantiating the necessary
    classes and executing them) but to call an already
    running app at each invocation (so, getting request
    and invoking, in a manner that I don't know, the
    running application).The container manages this. If you have data that must remain loaded, you can associate it with the class (use a static modifier.) This will complicate threading, however.
    I can use Axis to get the request, but it also grants
    me that my app will always be active?I think you are just using the wrong terminology here. What I think you are asking is whether the resources will be loded into memory at all times. If you want to ensure this behavior, you need to associate the data with a class. I'm not 100% positive, but I don't think Tomcat will unload classes in normal circumstances.

  • Seeking advice for timer design

    Hi,
    I am designing a timer application with Flex Builder 3 which
    is basically a stop watch that will remember when it was started if
    the page is reloaded (on a different computer, for instance). My
    original plan was to use JavaScript and mySQL but I'd like to give
    it a try with Flex first.
    Since Flex doesn't interact directly with databases I'm
    wondering if using mySQL to store the timer settings is the best
    idea? I think I could manage interfacing it with PHP but I'm not
    sure how much trouble this is going to be since I'm a new Flex
    user.
    How would you approach the project?
    Thanks,
    tekfetish

    Interfacing with a server part is very simple using Flex, PHP
    would be just fine - as well as any other server side
    language/system.
    Here is an Actionscript idea:
    You just create a HTTPService variable
    quote:
    * gateway : this is the communication layer with the server
    side php code
    private var gateway:HTTPService = new HTTPService();
    Then set properties of this service
    quote:
    gateway.url = '
    http://www.myServer.com/myScript.php';
    gateway.method = "POST";
    gateway.useProxy = false;
    If you'd like to pass parameters, use something like this:
    quote:
    parameters['someNameHere'] ='John Doe';
    parameters['whichTimer'] = 'timer 1';
    gateway.request = parameters;
    Then, define which functions should be called once the
    HTTPService returns [resultHandler on success and faultHandler on
    error:
    quote:
    gateway.addEventListener(ResultEvent.RESULT, resultHandler);
    gateway.addEventListener(FaultEvent.FAULT, faultHandler);
    Then do something in these functions:
    quote:
    public function resultHandler(e:ResultEvent):void
    // your code here...
    public function faultHandler(e:FaultEvent):void
    var errorMessage:String = "Connection error: " +
    e.fault.faultString;
    if (e.fault.faultDetail)
    errorMessage += "\n\nAdditional detail: " +
    e.fault.faultDetail;
    Alert.show(errorMessage);
    Finally, perform your request:
    quote:
    gateway.send();

  • Seeking advice ABAP With BW

    Hi Friends/Professional,
    I am working as ABAP Consultant, i would like to learn new technology in sap, i am seeking advice please help me.
    As a abaper shall i select BW or XI or ABAP HR
    which is best acording to industry india/us market.
    i am not interested in coding, please advice,
    few friends giving advice BW.
    shall i go BW or any other module.
    please give advice asap.
    Thanks in advance
    Manisha

    Hi,
    If ur good at ABAP, learn BI/BW, it is the best combination to prefer. Usaully ABAP''ers learn BW very fast( seen this practically). so go ahead & learn BI/BW.
    The  BI BOOKs /Material's which is provided bySAP:
    TBW10 (BI – Enterprise Data Warehousing)
    TBW20 (BI – Reporting & Query Design)
    TBW41 (BI – Extraction with UDI & XI and APD)
    TBW42 (BI – Advanced Enterprise Data Warehousing & Reporting)
    TBW45 (BI – Integrated Planning
    For syllabus details....check the following link...
    http://www.sap.com/services/education/certification/certroles/certificationtest.epx?context=FFC760B8923D16BB5150DAE63E7C1A6B331AF0B9E3A8F73CE3A9B7046E051044503600C911DBA13DCE978D3AC9057626D2B68111A7CD2D707E2EEC31213097E46EB790DD0106435EE0756F7B22F3FA4B4FF0645C06954BF3A150E023B4164DA2C3943DE02E599735DB9B4334B30B38FF20A1DC779D8F55E5F7A6893BDBFA38B94CF455E2A3E0E6851014966C90C80E173937CF7C2372A6FC%7cDA891B3C877030D9765B85CAE6AC82FC3EB6BC7DA10B7335
    Hope this helps,
    regards
    CSM Reddy

  • Seeking advice: moving universe between systems

    Hi all,
    I've been spending some time trying to work through the Business Objects documentation to try and understand what's involved in migrating a Universe from a non-SQL Server based system to a SQL Server based system.
    I'm seeking advice not having used Business Objects as I am attempting to estimate the work involved for a customer who currently uses BO. I don't know whether it's just a simple exercise of changing the database connection, or whether more work is involved. I've tried to get this information through SAP formally, but not currently being a BO customer they've refused to answer any questions.
    During the course of the migration of the underlying data tables, some columns in some tables have changed their data-type (e.g., from smalldatetime to datetime, or changes in length/precision). I'm also not sure if the tables have changed name: in the old system, they are named in a SQL type manner (i.e. <prefix>.<tablename>) however in SQL Server the objects will be placed in a database named <prefix> with each table being named <tablename>.
    Would this potentially require a full drop of every object in the Universe, or could this exercise be achieved without having to remove and re-add any objects by saying "this object in the Universe is now this object in the database".
    Any advice welcome.
    Cheers,
    Chris...

    Hi Chris,
    The universe is quite a simple tool in that it maps the sql directly to objects. So if the universe points to a table name that is no longer present, you wil get an error in the report, or when you check the integrity of the universe (tools, check integrity)
    The standard formats will move accross without an issue, however it is possible to use vendor specific code in objects. So for instance I could write
    nvl(my_table.my_col, 0) in Oracle
    or
    IsNull(my_table.my_col, 0) in MS SQL.
    If you objects contain these defintions then your job is going to get harder.
    Regards
    Alan

  • Seeking advice on SAP/cloud

    Hello all,
    Will be doing proof-of-concept on SAP on cloud, installing ERP and seeking advice/feedback.
    Based on the criteria of easier installation and faster delivery (and of course success story), what will be the recommended combination of platform (LINUX, Windows) and database (Oracle, MS SQL, MAXDB, DB2).
    Thanks in advance.
    Regards,
    Terry

    There are lots of  documents in SDN and other forums which will give much better idea ... but these are my observation
    Installation - Once you have gone past the access process to the cloud environment, after that you can hardly differentiate whether you are on doing a install on cloud or non-cloud. Though definately the time to get the infrastructure for install is much faster.
    Windows or Linux - I would put it as windows or Unix, not sure how many providers offer other flavors of Unix apart from Linux. But obviously the Product itself to some extent will govern what you choose, as has been the case in non-cloud env.
    Database - Again have seen Oracle and DB2 being mostly used. Not sure why, may be stuff like doing system refresh in 15 min ,  etc  are easier to achieve with products like Oracle and DB2.
    Regards,
    Neel
    I dont have any exposure to windows on cloud. So i am not aware of specific positives it might have in cloud.
    Edited by: Neelabha Banerjee on Nov 22, 2011 3:40 PM

  • Seeking advice for backing up Xserve G5 w/ RAID PCI card

    Hello all!
    I'm a newbie to Macs and server admin, and I have inherited the job of setting up a server at work for file storage. I'll do my best to give a concise description of our set-up - I'm looking for some advice on the last few odds and ends... mainly how we can backup the system.
    We bought an Xserve G5 with an option RAID PCI card. We have 3 500GB drives in the Xserve, configured to RAID 5 (giving us effectively 1TB of storage space). We will be using the server for data storage. About 20 computers will access the server over a network; we are using an assortment of Macs and PCS.
    I am seeking advice on backup systems for the server. In the event that the RAID5 fails, we don't want to lose our data. We just need a snapshot of the server; we don't need to archive data or to take the HD offsite. Our old server just used Retrospect to run incremental backups every night (with a complete clean backup once a month). Our current thought is to attach large external hard drive to our admin computer (not the server directly) and run nightly backups as before.
    The major points I have are:
    -Any thoughts on reliable 1 TB external drives?
    -Any recommendations on software that can backup from a RAID over a network? I found info for Retrospect Server 6.0 - it seems to do what we want, but is rather pricey.
    Thanks in advance for any advice! I really appreciate it!
    Xserve G5 Mac OS X (10.4.2)

    Greetings-
    We all started out as newbies at one time or another-no worries. That is why we are here.
    I personally use the Lacie branded drives. They are sturdy and reliable. My only thoughts here are to have at least three external drives-one for today's backup, one for a backup stored securely nearby, and one to be securely stored off-site. Rotate the drives daily so that your worst-case scenario will be a catastrophe that requires you to use a three-day old backup, depending upon how old your off-site backup is. Not ideal but better than the alternative. External drives are cheap enough these days to allow you to do this at a reasonable cost. Plus it is easy enough to throw one in a briefcase and tote it home for safety (just don't lose it!)
    I would stay away from Retrospect. If you search these forums you will find several instances of folks having some serious issues with the program. I use a program called Carbon Copy Cloner that does the job nicely for my basement server. There are ways to do the backups via the command line interface as well but I am not so familiar with those commands. You may have to dig a little deeper to find something that works for you.
    One of the other advantages of the FW external drive is that you can share it with other users, so perhaps you can set things up to have your network backups to go to that drive. Tis a thought.
    Luck-
    -DaddyPaycheck

  • Seeking advice on moving up to HDV editing

    I appreciate folks on this forum who have assisted me in the past when I’ve had questions about iMovie, editing, burning DVDs etc. Karsten, Lennart, Bengt and others from around the world have been very kind and helpful. I am now seeking advice from anyone regarding “moving up” to the HDV world of video editing.
    I recently "retired" my 2003 iMac G4 800 MHz flat screen and replaced it with an iMac 2.8 GHz Intel core 2 duo with a 320 GB hardrive and 4 GB of upgraded RAM. I have a 250 GB external drive as well. I also replaced my old Sony TRV-30 digital camcorder (a great old workhorse which is still going strong after 8 years) with a Canon Vixia HV40 that has high definition recording capability (uses mini DV tape).
    I had been using iMovie HD 06 to edit my video projects along with iDVD and Toast 6 to burn them to DVD. I would like to start shooting HDV video with the new Canon, edit and burn the projects to Blu Ray high definition discs. MCE Technologies (among others, I believe) offer both internal and external Blu Ray burners. I don’t believe that the internal Blu Ray burner would work in my new iMac, but I may be wrong on that. (Does anyone know the answer?)
    I believe that the MCE external Blu Ray burner would work with my equipment if attached to one of the Firewire ports or possibly through one of the USB 2 ports. (there is both a 400 and 800 Firewire port on this iMac model.) I was planning to acquire Final Cut Express as the editing program with Toast 10 Pro (plus the Blu Ray plug in) to burn the edited projects.
    Will this work? Is there anything I should know before I plunge “headlong” into HDV editing? Any advice would be most welcome. Thanks in advance.
    Message was edited by: scooper

    scooper wrote:
    .. iMac 2.8 GHz Intel core 2 duo with a 320 GB hardrive and 4 GB of upgraded RAM. I have a 250 GB external drive .. Canon Vixia HV40 .. external Blu Ray burner .. Final Cut Express .. Toast 10 Pro (plus the Blu Ray plug in)
    Will this work?
    I'm jealous.. wow! MY set-up..
    HDV is supported by IMHD6.. just to mention that. so, HDV-editing is same as SD..
    iMHD6 Export 'FullQuality' creates a file, which is HiDef too, for your Toast/BR workflow ..
    HDV creates on import (IMHD6/FCE) huuuuuge files, 40-60GBs/h .. 250 extHDD = 6h, full..
    actual price for extHDD with 1TB <100€ ..
    keep in mind: internal harddrive has to offer any time 10-20GBs free..
    not every BR-standalone-player accepts homebrewn disks..don't be disappointed...
    Toast and iDVD can burn DVDs (=which offer SD only) from HiDef projects.. no need to 'convert'.. both apps downsize automatically...
    FCE has a (very) steem learning curve.. totally different to iMovie.. no one-click-only, no templates, no ready-mades.. but zillions of options, 101% control.. (simple example: ken burns with zoom-to-zil and rotation and moving out of frame...)
    buy a book of FC Mastermind and AppleDiscussions-Level5 supporter *Tom Wolsky* - he's a sensei, in-depth-knowledge, hands-on (=no tech blabla), easy to read and understand.
    don't start with a project, start with 'tests for the trashbin'..

  • Need advice for future design and hardware I should purchase now.

    I was wondering if someone could assist me in making a decision on where I should take the future design of my network. Currently the design has no redundancy and I've been told recently that my company has a bit of money to spend and that it needs to spend it within the next 2 weeks. I am fairly new to the company so haven't been able to really think about future redundant designs nor have I studied much about designing networks. There are about 200-300 people that may be using my network at once, all users aren't at the location in question but they may be asking for resources from our servers.
    I've included a basic design of the "core" of my network and would like any suggestions for creating redundancy in the future and also optimizing the way data travels to our servers. Do people generally have redundant Layer 3 switches for the core in small networks such as mine? I will be replacing the 2811 since it only has 100Mbps connections and was thinking, perhaps replace this with a Layer 3 switch with the plan to have another identical Layer 3 switch installed to offer redundancy in the future.
    Also, would it be a good idea to move the servers into a vlan on the core? Thanks for any advice on what I should be purchasing now with plans for redundancy being implemented over a year.  -Mark

    40k Can go pretty quick depending on the scope. Your server farm especially should be dual-homed capable of surviving link, hardware, routing, and software failure.
    It's going to be best practice your server farm be in it's logical subnet, so failover mechanism can be controlled routing protocols, as opposed to FHRP's such as HSRP/VRRP/GLBP. Especially since you adjust routing timers to sub-second convergence.
    Budget will be the primary limitation (as it always it) but ideally dual 6500's running VSS and FWSM would be the ideal way. Data centers should be designed with high availability in mind, hence the need to 2x devices.
    Depending on the size of the SAN/Virtual infrastructure Nexus might want to be considered but you will chew up 40k before you know it.
    Also make sure the server farm is scaled properly. Your server farms should be oversubscribed in a much higher ratio compared to your access layer.
    CCNP, CCIP, CCDP, CCNA: Security/Wireless
    Blog: http://ccie-or-null.net/

  • Seeking advice: project sharing

    For five years, I've produced a quarterly newsletter (for teh day job) that averages 28 pages. In 2014, I'm preparing to change the project by lowering print run and page count and making it monthly - so we'll be producing the same number of pages in a year, but on a radically different schedule.
    To help me accomplish that, another employee will begin contributing content, including page layout. She uses our same CS5 license in the office next to mine and is comfortable but not confident (i.e. will be a full but junior partner for the first six issues while her Adobe chops catch up with her composition & design skill).
    So the question, broadly, is: what advice would you give in this situation? The coworker and I have done many prjects together over the years, and I'm not at all worried about issues of ego or control. But I've been making this product myself, and have a lot of "shorthands" built in to my workflow that she doesn't necessarily have to emulate. Also, because I've answered to no one for so long, I expect some headbutting as we learn from each other - and might have use for robust version control, so we can see different principles in action on a page.
    Does it make sense to break out InCopy (which we don't currently own) for such a short-distance working relationship? Is there a plug-in by which she could "check out" pages 3 and 4 from my working master, in the way that a Dreamweaver server can handle different versions of an HTML page?

    Is this about Indesign or Dreamweaver?
    Just wondering: so much effort for a small news letter that published only four times a year? Two people, optimised workflow, shorthands, version control?
    I'm sharing my documents with a colleague who is translating them to Chinese inside Indesign. There is no other way than sending her all the INDD and footage, so she can work.
    Instructions about changes have to be given. I see no way to make this more efficient and actually, there's no need to.

  • Seeking advice for a self-teaching programme of new Java techs

    Hello,
    after a few years of technically poor projects (I learnt a lot in terms of process, quality, and non-Java technologies, by the way), I'd like to catch up with mainstream Java development.
    I am preparing a programme to learn a few technologies I have missed over the past years. I plan to develop a couple of utilities (simple CRUD apps) on my spare time, gradually including bits and bits of the bricks I want to grasp.
    I need some advice though to make this learning effective.
    My first priority is to invest in reusable knowledge, so I will stick to "standard" bricks, and will aim mostly at off-the-shelf tools that work out of the box
    (for example, I don't want to spend hours to fix a DB install&configuration issue, as I know I will probably never be paid for that kind of job).
    Specifically, the technologies I plan to embrace are:
    * Java SE 5 and 6
    * jUnit 4
    * Spring
    * EJB 3 & JPA
    * Web UIs
    Here are the points where you might help me:
    * JDK 1.5 vs 1.6
    No real doubt here, the biggest language gap was introduced by 1.5. If I'm correct, 1.6 essentially brings new APIs; as long as those APIs are not my primary target (I'll try them only after I've completed my first round of catch-up :o), I am not missing much language-wise, am I correct?
    I could also jump directly to 1.6, ignoring its new APIs, and concentrate on
    1.5 language features, but the risk is stability (losing time with a 1.6 beta bug).
    * jUnit 4
    Nothing special, except I have practically read nothing about it on these forums (contrast this with TSS, where there are regular threads disputing the relative merits of jUnit vs TestNG).
    Is v4.0 a non-event? It seems to bring some niceties (mark test components using annotations), but how much comfort did it bring to those of you who have used it over jUnit 3.8.1?
    * Spring
    Leaving aside that it's on my must-do list, I actually wonder which version I should try?
    I also feel I should try it first on a non-EJB app, to avoid mixing concerns. I will skip the JDBC wrapping API though, as I already have a lot to do with other persistance APIs.
    Any feedback?
    * EJB3/JPA
    (I formerly worked with generation 2.1 of the EE specs)
    1) The biggest issue here is to find a reliable EJB3 container.
    * I've heard JBoss is EJB3-ready, but I don't think it's certified yet.
    * Project GlassFish (Sun-driven open-source JEE5 container) is still in the making
    * Sun SASP9 is, well, released, but I don't know how much of JEE it supports, and haven't investigated into the licensing details yet
    Any feedback on what you're using as a JEE5 container?
    2) As far as EJB vs JPA goes, I also have the plan to use the persistence API
    outside of EJB (likely, I will develop the same simple CRUD app twice,
    once as a Webapp and once as as Swing app, both over the same DB schema).
    But maybe this is pointless, as the entity annotations and API are agnostic; have you experienced some difference that deserve thorough learning?
    3) Obviously I will need a DB. If the EJB container includes one, fine, at least for the EJB part, otherwise I'll need to install and configure one, that will need to run on a desktop PC on MS WIndows, and preferrably include a handy SQL client console for manual verification.
    I know enough of SQL to build even a sub-optimal application schema. However I know nothing of a DB administration, and I don't want to invest in that direction.
    Any advice as to a simple off-the-shelf free DB that runs on Windows?
    * Web UIs
    The dilmena here is two-fold:
    1) In term of "view" technology, I hesitate between plain JSP and JSF
    If I understand correctly, JSF is a specification for reusable view components.
    My local job market leaves me no oportunity to be a UI component developer, whereas I am more likely to someday be tasked to integrate 3rd-party UI components into a business project.
    Is my surface understanding of JSF correct? In this case, how much of JSF should I know if I don't have to develop JSF components?
    2) In terms of controller, as part of my Spring learning, I'll probably have a look into Spring MVC.
    My question is, how much of the servlet/JSP API does springMVC show/hide?
    Is it a good idea to use springMVC to learn servlets and JSPs, or should I stick to manual servlet+JSP?
    I should add that I've worked with servlets/JSPs formerly (as of J2EE 2 or 3),
    I only need to dig new servlet/JSP features.

    Jim_Found wrote:
    okay, so normally you would write the following if-elseif block in order to achieve this
    if (MessageType.TYPE_A.equals(myMessageType)) {
    // do this
    } else if (MessageType.TYPE_B.equals(myMessageType)) {
    // do that
    } else if ...
    but, if you use <this design pattern> and create this <class>, you can suppress the above block to
    CoolClass coolObj = new CoolClass();
    coolObj.doMagic(myMessage);Funny enough inside the doMagic() method would be something along the lines of
    if (MessageType.TYPE_A.equals(myMessageType)) {
    // do this
    } else if (MessageType.TYPE_B.equals(myMessageType)) {
    // do that
    } else if ...Mel

  • Seeking advice on Best Practices for XML Storage Options - XMLTYPE

    Sparc64
    11.2.0.2
    During OOW12 I tried to attend every xml session I could. There was one where a Mr. Drake was explaining something about not using clob
    as an attribute to storing the xml and that "it will break your application."
    We're moving forward with storing the industry standard invoice in an xmltype column, but Im not concerned that our table definition is not what was advised:
    --i've dummied this down to protect company assets
      CREATE TABLE "INVOICE_DOC"
       (     "INVOICE_ID" NUMBER NOT NULL ENABLE,
         "DOC" "SYS"."XMLTYPE"  NOT NULL ENABLE,
         "VERSION" VARCHAR2(256) NOT NULL ENABLE,
         "STATUS" VARCHAR2(256),
         "STATE" VARCHAR2(256),
         "USER_ID" VARCHAR2(256),
         "APP_ID" VARCHAR2(256),
         "INSERT_TS" TIMESTAMP (6) WITH LOCAL TIME ZONE,
         "UPDATE_TS" TIMESTAMP (6) WITH LOCAL TIME ZONE,
          CONSTRAINT "FK_####_DOC_INV_ID" FOREIGN KEY ("INVOICE_ID")
                 REFERENCES "INVOICE_LO" ("INVOICE_ID") ENABLE
       ) SEGMENT CREATION IMMEDIATE
    INITRANS 20  
    TABLESPACE "####_####_DATA"
           XMLTYPE COLUMN "DOC" STORE AS BASICFILE CLOB  (
      TABLESPACE "####_####_DATA"  XMLTYPE COLUMN "DOC" STORE AS BASICFILE CLOB  (
      TABLESPACE "####_####_DATA" ENABLE STORAGE IN ROW CHUNK 16384 RETENTION
      NOCACHE LOGGING
      STORAGE(INITIAL 81920 NEXT 1048576 MINEXTENTS 1 MAXEXTENTS 2147483645
      PCTINCREASE 0 FREELISTS 1 FREELIST GROUPS 1 BUFFER_POOL DEFAULT FLASH_CACHE DEFAULT CELL_FLASH_CACHE DEFAULT))
    XMLSCHEMA "http://mycompanynamehere.com/xdb/Invoice###.xsd" ELEMENT "Invoice" ID #####"
    {code}
    What is a best practice for this type of table?  Yes, we intend on registering the schema against an xsd.
    Any help/advice would be appreciated.
    -abe                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               

    Hi,
    I suggest you read this paper : Oracle XML DB : Choosing the Best XMLType Storage Option for Your Use Case
    It is available on the XML DB home page along with other documents you may be interested in.
    To sum up, the storage method you need depends on the requirement, i.e. how XML data is accessed.
    There was one where a Mr. Drake was explaining something about not using clob as an attribute to storing the xml and that "it will break your application."I think the message Mark Drake wanted to convey is that CLOB storage is now deprecated and shouldn't be used anymore (though still supported for backward compatibility).
    The default XMLType storage starting with version 11.2.0.2 is now Binary XML, a posted-parsed binary format that optimizes both storage size and data access (via XQuery), so you should at least use it instead of the BASICFILE CLOB.
    Schema-based Binary XML is also available, it adds another layer of "awareness" for Oracle to manage instance documents.
    To use this feature, the XML schema must be registered with "options => dbms_xmlschema.REGISTER_BINARYXML".
    The other common approach for schema-based XML is Object-Relational storage.
    BTW... you may want to post here next time, in the dedicated forum : {forum:id=34}
    Mark Drake is one of the regular user, along with Marco Gralike you've probably seen too at OOW.
    Edited by: odie_63 on 18 oct. 2012 21:55

  • Mac Pro buying advice for Graphic Design

    I am going to be purchasing a Mac Pro and was hoping I could get some advice on the right set-up for what I do. I use CS4 heavily on the Mac for graphic design spending a good amount of time in Photoshop working with files that can easily go over a few gigs (when working on convention booth displays, etc.). I also use my Mac for casual video editing in iMovie, and use Apple's other great offerings like iWeb, iPhoto, etc.
    My main questions are:
    - I'm leaning towards the Quad. I keep my Macs for at least 5 years. Is that a good decision?
    - If I do go with the Quad should I opt for the 2.66 or 2.93?
    - At the moment I am running 8 Gig on my current Mac, should I configure this new one with 8 Gig?
    - Is the Radeon the correct card choice for a heavy Photoshop user?
    I hope these questions aren't ones that have been asked a thousand times. It's a lot of money to spend (especially right now) and I want to make sure I configure it correctly for what I do.
    Thank you in advance for your input.

    Hi hatter,
    Thank you again for your responses. That really helps clear up the RAM situation. I think that is part of what swayed me towards the 8-Core as opposed to the Quad. Please correct me if I am wrong, but looking at prices of RAM on Crucial I see that if I wanted today to put 16 Gigs of RAM in a Quad-Core it would cost me $1200 (for 4 4-Gig chips using all slots). However if I was going to put 16 in the 8-Core I could do it for $400 (8 2-Gig chips using all slots. Actually it would only be $200 more since I configured my Mac with 4 2-Gig chips from Apple for an extra $100). Seems like having more slots open will save me money when I want to upgrade ram since I don't have to go for the 4-Gig chips. That coupled with the theory that Snow Leopard is around the corner and will take advantage of the 8-Core it seems like opting for a low end 8-Core vs a high end Quad is worth the $300 difference.
    Side note: Are there any reports that say running more 2GB chips is worse than running fewer 4GB chips? Just curious.
    Even though I understand that the processor speed makes a difference I am hoping that the advantages of more RAM slots and future benefits of Snow Leopard, CS5 and other apps that will take advantage of the 8-Core in the near future will make it the right decision. If I could afford a faster 8-Core obviously I would opt for that but with a jump of $1,400 to the next processor speed I am afraid it is out of my reach.
    Regardless I'm sure I will feel very spoiled as soon as I unwrap this beast and get it up and running. Then in a year I'll give into the temptation of slipping a SSD in for the boot drive and I'll have to find a way to contain my excitement. I have a SSD on my laptop and WOW what an amazing difference it really is.
    Wow, these forums are great. I can't believe how helpful all of your responses have been. I feel much more informed and was able to order my Mac today not feeling like I was shopping in the dark.
    Thank you all again!

  • Seeking advice on my first build, $3000 budget, please critique. Thanks!

    Hey all,
    Been lurking around here for some time now and I must say it's been interesting. A bit of background before we get to the good part: I'm a freelance website designer (one that does a lot of graphic design), so my workhorses are Ps and Ai, but lately I've been getting more and more freelance film work. I've reached a crossroads as my current system struggles to handle the load in Pr, especially in AE. I've decided to build a kickass system that will hopefully improve my workflow enough to take me more down the film path (in addition to dominating my Ps and Ai needs). This will be my first build, so all advice is welcome and much appreciated. I've tried pretty hard to do some research before posting, and here's what I've come up with:
    The Build:
    Category
    My Choice
    Price($USD)
    Comments
    CPU
    Intel Core i7-3930K Sandy Bridge-E 3.2GHz (3.8GHz Turbo) LGA 2011 130W Six-Core Desktop Processor
    569.99
    Is it worth waiting until the Ivy Bridge 4930K comes out? Looks like that won't be till the fall..
    GPU
    EVGA 04G-P4-3687-KR GeForce GTX 680 FTW+ w/Backplate 4GB 256-bit GDDR5 PCI Express 3.0 x16 HDCP Ready SLI Support Video Card
    589.99
    Would you choose the 680 over the 580?
    mobo
    ASUS Sabertooth X79 LGA 2011 Intel X79 SATA 6Gb/s USB 3.0 ATX Intel Motherboard
    324.99
    I honestly know nothing about mobos.. should I be concerned that this is from 2011? Is this overkill?
    RAM
    Kingston 64GB (4 x 16GB) 240-Pin DDR3 SDRAM DDR3 1600 ECC Registered Server Memory DR x4 Model
    519.99
    Is Kingston a good brand? This seems like quite a deal.
    disks
    1x Samsung Pro 840 series 128GB SSD, 2x Seagate Barracuda 7200.14 3TB 7200 RPM RPM 64MB Cache SATA 6.0Gb/s 3.5" Internal Hard Drive
    420.97
    SSD for OS&programs, 1HDD for media, 1HDD for cache and exports. This is the area I'm thinking I might improve on as I'm a bit under budget. What additional disks/purposes do you recommend? Seems like I shouldn't bother with RAID unless I get more disks.
    PSU
    CORSAIR HX Series HX750 750W ATX12V 2.3 / EPS12V 2.91 SLI Ready CrossFire Ready 80 PLUS GOLD Certified Modular Active PFC
    129.99
    Seems like a good PSU/cooling unit. Don't know much here though.
    Case
    Fractal Design Define XL R2 FD-CA-DEF-XL-R2-BL Black Pearl Steel ATX Full Tower Computer Case
    149.99
    I like that this has 2 audio jacks. Has 8 internal 3.5'' drive bays, 4 external 5.25'' drive bays, and 9 "expansion slots", whatever those are. It does not mention 2.5'' bays, so I assume it has none. Does that mean I won't be able to mount my SSD? Should I go with a different case for that reason?
    Total:
    2689.91                                                                  
    So I'm a bit under budget. What do you think would be the best components to upgrade?
    I'm thinking I might get 3 more Barracuda 3TBs (or maybe smaller to cut the cost a bit), put 4 of them in RAID3 for media, keep the leftover drive for exports and scratch. Does that sound like a good idea? If I did this, I think I would also get this controller: Areca ARC-1213-4X PCI-Express 2.0 x8 Low Profile SATA / SAS 4-Port PCIe 2.0 External SAS/SATA RAID Controller. Coming in at 389.99, it's not cheap, but I am willing to go a bit over budget if it will make a big enough difference. With this RAID setup, I'd be looking at around $3300, which is pretty much my max. Thanks in advance for your help!!
    EDIT: Forgot to mention that I will be running CS5.5, probably going to go for the cloud upon release of CS???

    Mkudisch, it looks like a great PC you're going to build, it looks pretty much like the one i've built a couple months ago, I'm not a master pc builder but I would like to give you a little advice about the componenets:
    CPU: Great choice, I have the same cpu.
    GPU: it looks like a great gpu, just one thing if you plan to overclock it you should buy one with better airflow (at least 2 fans), I have an asus gtx 670 4 gb.
    MOBO: great choice, probably you already now it, it is a 2011 model but if your saying it because the name says LGA 2011 it means that it is socket 2011 the one for sandy bridge-e cpus, (I have a P9X79 WS mobo)
    RAM: probably I'm wrong but as far as I know all Asus X79 mobos will only take 8gbs sticks per slot and not 16 gbs ones, you should probably look for 8x8gbs quad channel, I have G.SKILL Ripjaws Z Series 8X8gbs and they're great, if you want search for any other brand model but first check if you mobo will take 16gbs sticks per slot.
    Disks: great SSD, I've heard a lot of great things about samsung 840 PRO.
    PSU: I think you should consider at least a 1000W PSU, in case you want to add another gpu or more hdds in the future you will be already covered, Corsair it's a great brand also stay away from any Rosewill PSUs I bought a 1300w lightning series and that piece of crap exploded 30 seconds after turning it on.
    Case: don't know about that case but it looks like a great choice I have a HAF X 942.
    Also you forgot about a CPU cooler, there's a lot of great cpu coolers, I would recommend you Noctua NH-D14 SE2011 that's the one I have (and it's great if you plan to overclock you i7-3930K) just make sure the cpu cooler you buy is compatible with you cpu, if you buy Noctua NH-D14 SE2011 make sure it will fit in you case because that cooler is HUGE!.
    Well that's my advice like I told you before I'm not an expert or anything just another guy like you who built a PC similar to the one you want to build.

  • Zones Designing in firewall

    Hello Experts
    We have core firewall in datacenter. We have web servers (front end), application servers (middle end) and database servers (backupend) and backup appliances. The OS is a mix of windows/linux
    What is the best practice to design the zones in this enviornment for good security. I means lets say putting all web servers in one zone, application servers in second zone and database server in another zone and backup appliances in sepearte zone?
    1- But what about security of say application to application servers?
    2- And sometimes I heard application to database does not like firewall?
    3- Similary backup appliance to servers huge traffic passing through firewall?
    4- Also OS is of different type. Should I also consider to put same OS in same zone?
    Appreciate the input

    It would be best if you were able to separtate all services into seperate zones, but this isn't always possible.  I would suggest putting all frontend servers in one zone,  and all other servers (backend, database, and backup servers) in a second zone.  Then restrict required access by using ACLs and make sure that traffic is explicitly permited, do not allow permit IP any any or even permit IP any to server IP.  Specify which ports should be allowed to access the various servers.
    Please remember to rate and select a correct answer

Maybe you are looking for