Want to know The Truth About CPM?
Showing posts with label Windows 2008. Show all posts
Showing posts with label Windows 2008. Show all posts

27 November 2011

AWS tricks #2 -- Treating that AWS instance as a real server


Introduction

I really like Amazon Web Services for development purposes.  It’s fast (enough), big (more than enough), and flexible (ah, just right) for my purposes.  What do you do with it?  Have you given any thought to treating that AWS EC2 instance (that’s an Amazon Web Services Elastic Computing Cloud virtual machine for the uninitiated) as a full-fledged server?

Splendid isolation

One approach, per the my earlier post, is to install all client software, including Microsoft Excel, on the server and just use Remote Desktop Connection (RDC) to connect to that EPM cloud instance.  This is a great approach if you are the only user of the instance (like me in my development/kick the tires/play with the product till you understand it usage).  But what happens when you want to collaborate with someone else or have multiple users check out the awesomeness that is your Essbase/Planning/HFM/Profitability/Close Management/Profitability/Studio/ODI/etc./etc./etc. work?

No geek is an island

Amazon Web Services (AWS) is predicated on servers, and after all the EPM software is server-based – why not treat your instance like a server and your PC like a client?  While connecting as a client to your EPM instance in the cloud is slightly harder than connecting to a server down the hall, it is easier than you think and requires just a few changes to the AWS Security Group.

 

And of course there is a huge benefit – more than one person, e.g., your customers, be they internal or external, can connect to the EPM server.  You do like to share, right?

Requirements

You must have:

  • A valid AWS id
  • An already created security group
  • A PC with Excel 2007 or 2010 (2003 is also supported but the examples show 2007/2010’s ribbon interface)
  • Already read the most-read (Google Analytics tells all) post I have ever written
  • Microsoft’s Remote Desktop Connection (RDC).  This client is standard in Windows; Macintosh users go here:  http://www.microsoft.com/mac/remote-desktop-client)
  • Time and a willingness to experiment  :)

 

Make sure you do not have Smart View installed as its download and installation is covered below.

Understanding AWS Security Groups

Ports

Security groups are used to define the open ports between an Amazon Web Services (AWS) instance and its clients.  By default all server ports are closed by Amazon’s firewall.  These ports must be opened for a client to connect to the instance.

 

As was illustrated in the “Yeah, yeah 11.1.2.1 is here, so what?” instructions, managing the AWS Windows instance required port 3389 to be opened to allow RDC to connect.

IP addresses

AWS provides a powerful method to limit unauthorized connection to an instance – internet protocol (IP) address filtering.  In essence, AWS allows you to do define which client external IP addresses will connect to a given instance.  It can be the whole world, or it can be just the EPM users in a given location or company.

 

This external IP address is not the same as your PC’s internal address on the LAN – it is the address the outside world sees and interacts with; usually one address is shared for a given organization or location.  Sophisticated AWS security is beyond the scope of this post (not to mention my abilities) but to get a taste of what is possible, investigate Amazon’s Virtual Private Cloud and the Elastic IP component of Elastic Compute Cloud for more information. 

 

For simplicity, this document will use a 0.0.0.0 IP address, as did the 11.1.2.1 instructions.  In AWS, a source IP address of 0.0.0.0 allows connectivity from and to all client IP addresses on the internet, including the ones belonging to those nasty people who do nothing but troll the ‘net looking for vulnerable servers.  You do not want to use this completely open IP address range for anything other than training purposes.  Figuring out your public IP address is easy; here is one site that provides this free service:  http://whatismyipaddress.com/ip-lookup

 

One final note on wide open IP addresses -- when the authors connect to AWS, whether it be from work or home, they always use their public IP address, never 0.0.0.0. 

Opening up the ports

A good test of true client/server connectivity is Smart View against Essbase.  Smart View uses ports 19000 for shared connections and 13080 for Essbase private connections.  If you set up your security group as described in the post-I-am-mentioning-yet-again, the only open port in your AWS security group will be Remote Desktop’s 3389.

Modifying the Security Group

You will need to expand the open ports by first logging in to AWS’s Elastic Compute Cloud (EC2) console and then clicking on the Security Groups link: 

 


Once in the group, in this case it is called KScopeTest, click on the Inbound tab:


Adding port 19000

You will see port 3389 already open.  To add port 19000, type in the port number in the Port range field and then the Add Rule button:

Adding port 13080

Do the same for the private connections on port 13080:


You can click on the Apply Rule Changes button at any time to commit the new port number; you must do so at the end of the port addition steps.

Knowing when to connect

As you know, it can be hard to determine when an AWS instance has started up after launch from an Amazon Machine Image (AMI).  The AWS console instance information is limited to four states:  terminated, stopped, pending, and started.  Started means just what it says – the instance is started, but all that really means initially is that the server has been turned on, not that you can connect.

 

When an instance has not fully booted, it cannot accept connections; the only way to know if you can connect or not is to try.

Ping comes to the rescue

One way to gain some insight into whether an instance can be connected to is to see if basic networking protocols can be used, e.g, pinging the box using the ping utility.  However, by default AWS instances can not be pinged because the protocol that allows ping and traceroute is disabled.  Thus a further modification of the security group is required to allow  ping from your client to reach the AWS instance. 

 

To do this, click on the Create an new rule dropdown and select Custom ICMP rule:


 

When you select Custom ICMP rule, the Type dropdown automatically switches to All.

 

Click on the Add Rule button to add the ICMP ports.  You must also click on the Apply Rule Changes button to commit the port changes.


 

Once committed, the Security Group should look like the following:


Port review

Port

Purpose

ICMP/All

Allow ping, traceroute, etc.

3389

Allow Remote Desktop connection

13080

Smart View private connections to Essbase

19000

Smart View shared connections, Workspace, etc.

 

We have only opened the ports required for connection:  ICMP, 3389, 13080, and 19000. 

 

As an example of port access, if your EPM instance were to support the classic Essbase Excel add-in, add port 1423 to allow connecting to Essbase and ports 32768 through 32778 to allow connections to the Essbase agent and databases.

 

As a general rule of firewall security, only open the ports needed for your work.

Launch that instance

The detail and rationale behind the steps required to launch an instance were covered in the post-whose-name-must-not-be-mentioned-lest-we-all-go-mad; refer to it for background.  If you dare.

Which one?

For the purposes of this example, John Booth’s free and public EPM 11.1.2.1 AMI will be used.  You can find out all about the AMI at http://www.metavero.com.

3…2…1…launch!

To find and then launch that AMI, enter the AMI id ami-de2bd4b7 in the filtering textbox in the AMI section of the EC2 console and click on the Refresh button.

 

Once it appears, right click on the AMI name and select Launch Instance.   Make sure you use the Security Group KScopeTest you previously created.

 

 

Follow the Launch Instance wizard (see the below for the kinds of selections you should make) and at the Review step start the instance by clicking on the Launch button.


Two ways to see if the instance is really ready

Approach #1 – Check System Log

Right click on the instance and select “Get System Log”.


Review the log

Remember, you are firing up a Windows 2008 server.  It is not going to be fast.  Give yourself time to get a cup of coffee.  In fact, you can likely dump the grounds, scrub the filter basket, put in a new filter, measure out the coffee grounds, put in fresh water, and hit the start button and wait for that delicious pot of Life Blood.  In other words, there is likely a 10 to 15 minute wait ahead of you. 

 

You will eventually see something like the below; you may need to check back several times.  When you do see it, the log is telling you the truth – you can now connect via RDC.


Approach #2 – Ping the server

Checking the system log is all very well, but it can require multiple tests to ascertain if the instance is strated. 

 

Instead, the familiar ping command can be used to go against the instance – when Windows has booted, the ping command will be echoed back to your client.

Get the DNS name

Once the AMI is launched, grab the public DNS name by trying to connect to the Instance.  Yes, that instance will almost certainly not be ready to connect to – you are just going through the motions to get the address.

 

Right click on the instance name and select Connect.

 

The AWS console will show the DNS name.  Copy that to the clipboard.


Ping from the comand line

Start up a command window, and then enter ping DNSnameIn the case of the above DNS address (every launched instance will have a different one and yours will not match the above), the command would be ping ec2-184-73-120-11.compute-1.amazonaws.com

 

Although a successful reply from the instance is not complete proof that Windows is ready receive Remote Desktop connections, it is a good gauge that the EC2 instance and related Elastic Block Storage (EBS) drive have been created from the AMI and that the OS is in the process of booting up.


 

This is one of those differences between a GUI and a command line – the GUI requires you to act again and again, while the command line just keeps chugging away until it receives a reply.  Which would you rather do?

Connect via Remote Desktop

Once you get a reply from your ping command or from the Get System Log menu, connecti via Remote Desktop by going back to that AWS dialog box and then clicking on the Download shortcut file link.

 

In Firefox, Remote Desktop will open the .rdp (a config file for RDC) file.

 

A scary message will ensue.  Ignore it – this has nothing to do with AWS and is intead a feature of Windows 7/Remote Desktop.



Enter the password “epmtestdrive”.


 

Another scary message pops up – again this can be ignored by clicking on the Yes button.


 

When you have connected to your instance, minimize your Remote Desktop window as you will now connect from your client PC.

Connecting to an EC2 EPM image from your PC

Prove that Workspace is running

On your client PC, launch a browser instance (this example shows Internet Explorer as Workspace is optimized for that browser) and create a url to launch Workspace.

 

The normal Workspace url is:

http://servername:19000/workspace/index.jsp

 

The public DNS name of this instance is ec2-184-73-120-11.compute-1.amazonaws.com

 

Incorporate the DNS name into the url as follows (remember, in your case the DNS name will be whatever Amazon assigns to the instance):

http://ec2-184-73-120-11.compute-1.amazonaws.com:19000/workspace/index.jsp

 

Type that url into the browser’s address bar and hit the Enter key.  If all of the services are running on the AWS instance, you should see the below Workspace login screen.


 

Enter a username of admin and a password of epmtestdrive and you will see the familiar Workspace home page.


Download SmartView

The goal is to connect Smart View, not Workspace.  Navigate to the Workspace Tools menu, select Install, and then Smart View.  This will begin the download to your client PC.

 

NB – This guide assumes that Smart View is not installed.  If it is, unistall it to follow the remaining steps.

 

Internet Explorer will prompt for a save location for Smart View.exe.

Install Smart View

Once saved, navigate to the same location and right click on SmartView.exe and select Run as administrator.


 

Vista and Windows 7’s User Account Control will ask if you really want to run this file – you do.

 

Click on Run and the installation process will begin.


Pick a language – in the case of this guide, that is English (United States) and click on OK.  The installation process will begin.

 

The installation process begins…

 

Follow the wizard all the way through to the end.

 

Launch your PC’s local copy of Excel – you should see the Smart View ribbon.

Defining the Application Provider Services (APS) address

After going to the Smart View ribbon, click on the Open button and then the Shared Connections link in the Connection Manager.  Smart View will surface a dialog box asking you to modify the url for Shared Services.  Smart View connects to Essbase through APS – this is the address you are really providing.

 

In the case of this EPM instance, all services are installed on a single server.

 

 

The default url is:  http://server:port/workspace/Smart ViewProviders

 

Using the same DNS name that was used to ping the server and to connect to Workspace, change that URL to (again, your DNS name will differ from the below):

http://ec2-184-73-120-11.compute-1.amazonaws.com:19000/workspace/Smart ViewProviders

 

Paste it into the Shared Connections URL dialog box in Smart View’s Options dialog box.

 

Smart View will then show the Connect to Data Source dialog box.  Enter the same username and password of admin and epmtestdrive and then click on Connect.

 

Now go into Connection Manager and click on the server drop down to define an Essbase server.


 

Then click on the Add new server dropdown.


 

Paste in the public DNS name of: ec2-184-73-120-11.compute-1.amazonaws.com and click on OK.

 

Smart View will show you the Essbase service in your instance.  Below are the sample applications that come with Essbase.

 

 

From here on out, you can connect to Essbase just as if the server was in your data center.

 

It’s the same Essbase you know and love, only in the cloud.  It really is that easy.


When you are done

AWS instances are billed by the hour – if there is no usage, nor is it anticipated, it might be in your best interest to stop the server. 

 

If you want to retain your work you have two choices:

1)      Stop (not terminate) your server via the EC2 console.  This is akin to shutting down your server.  You can bring it up at any time and will incur small storage costs based on size of $0.10 per gigabyte month.

2)      Create an AMI (coming soon on this blog) from the instance and Terminate the instance.  Charges for AMI storage are slightly less than EC2’s EBS model, but mean that you have to launch the AMI and manage multiple AMI copies.

 

If this was throwaway work, simply terminate the instance.  Remember, there is no going back to the data once the instance has been terminated.

DNS changes

Once you stop (or create an AMI and terminate the source instance) the instance, the public DNS name will be lost.  On subsequent startup, you will need to rename Smart View’s APS server name. 

 

One way around this is to use AWS’ Elastic IP service; find out more about it here:  http://aws.amazon.com/ec2/  Elastic IP will allow you define a static IP address and AWS will map the dynamic IP address to that static one so that clients do not have to change their server names; there are further configuration steps beyond the scope of this guide – search the web for more information.

 

Two alternate DNS providers are:

http://www.no-ip.com/

http://www.dyndns.com/services/dns/dyndns/

Conclusion

What do you now have?

  • A fully functioning Essbase server in the cloud that you, and your colleagues, can use for any purpose Essbase might fill.
  • A test server that can be created in minutes for experimentation when your development server is unavailable.
  • A platform for learning the features of the very latest EPM release.
  • A server that can be almost instantly adjusted, whether it be the number of CPUs, the amount oF RAM, or the hard drive size.

 

All for mere pennies an hour.

 

 


24 October 2011

AWS tricks #1 -- How to get rid of terminated volumes

Introduction

I spent this weekend figuring out how to install ODI 11.1.1.5 using SQL Server 2008 not Oracle 11g as the database back end and I now know two things:

  1. I really ought to be doing this on Oracle instead of SQL Server as all of the blogs seem to cover the settings for Oracle.  I find it interesting that the 10.1.3.x blog posts seemed to be oriented around SQL Server and 11.1.1.x around Oracle.  However, as my environment was SQL Server, not Oracle, I was stuck with figuring it out.
  2. I am not very good at installations.   This may not come as a shock to anyone, especially me.  :)

In an upcoming blog post, once I am convinced that I have everything working tickety-boo, I will document the settings and required downloads to make ODI and ODI Studio work in a 64-bit Windows server context.

Putting aside the additional grey hairs and general agita installing ODI caused me, this exercise got me back into the Amazon Web Services (AWS) cloud and that was a life-saver.  Just like a virtual machine, I could spin up as many instances of the server as I needed to blunder my way through the install and could snapshot it when I had something halfway working.  Unlike a virtual machine running on my laptop, I had 17+ gigabytes of RAM, super-fast networking (you haven’t lived till you switch from DSL to whatever AWS runs – I was getting 39 megabit per second downloads from Oracle’s website), and real server power.  

However, as I was looking at my AWS account, I saw that I had a bunch of AMIs, snapshots, and volumes that I couldn’t really account for.  Oh, I created them all right, but I hadn’t used them in ages and I was getting billed each month for them.  A mass culling ensued but it occurred to me that this is  a good time for some basic definitions.

AWS definitions

One of the (many) things that confuse me about AWS is the concept of “volumes”.  Simply put, a volume is a hard drive.  

Then there are these things called “snapshots”.  Snapshots in AWS’ Elastic Block Storage (EBS) world are just what they sound like – snapshots of volumes that you can restore back to at any time.

It is very important to note that the hard drive of an Amazon Machine Image (AMI) (a predefined server you can start) is ephemeral, i.e., your laptop’s hard drive this is not – when the instance is killed, the hard drive is gone.  It’s very easy to blow away all of your work if you don’t understand how AWS treats volumes.

This post will cover what a volume is, and how, when, and why you would get rid of a volume (you get charged for each one) as they have a way of piling up when least expected.

What’s ephemeral, and what’s not

A really good explanation of AWS’ drive ephemerality can be found here:  http://shlomoswidler.com/2009/07/ec2-instance-life-cycle.html.  If you cannot be bothered to jump to a well written blog, the gist is that EBS-backed instances have volumes that persist as long as the instance is not terminated.  

And how does that affect you?

Well, when you stop an instance (Stopping an instance is akin to powering down a server) the volume is still extant.  It makes sense (to me at least) that a volume hangs around when a server instance is stopped.

When you terminate (Terminating an instance is like you shut down the server, ripped it out of your data center, and naively took it to a recycling center which then had the hardware shipped tout de suite to Liberia for completely unsafe disposal.) an instance, there is a chance the volume will still hang around.  But why is it still there when an instance is terminated?  Doesn’t that destruction of the instance sort of imply that the hard drive volume should be trashed as well?

Remember, if you started your stopped instance back up, whatever changes you made to your boot drive are still there.  

But what about the volume that is the product of a terminated instance?  If you reattached it to a new instance, are your changes there?  Nope, all of the changes you made are gone.

An example

Let’s pretend that you fired up John Booth’s EPM 11.1.2.1 AMI (go to http://www.metavero.com), did cool stuff, and then terminated the instance.  Guess what – that C: drive is still around, and you’re getting billed for it.

Not deleted till it’s detached

Here’s what the web console to my AWS account looks like with four volumes attached to stopped instances and one 100 gigabyte volume from a terminated server instance.  
What is that volume doing there?  Did I really terminate that instance?  

As I wrote, I’m being charged (not very much, but still) for storing that 100 gigabytes.  I want to make that unloved hard drive go away.

It’s dead, Jim

That instance is deader than dead.  That drive volume is in AWS purgatory.

What’s going on?

That volume is going to hang around till you delete it.

Why?  A little searching of the tells us that the AMI must be set up to delete the volume on termination.  If it’s not, then the volume must be manually deleted.

NB – A future blog post on launching an AMI from a command prompt will explain how to launch an AMI with that parameter; if the AMI has been set up to not delete the volume on termination, you must manually delete the volume as shown below.

How to delete a volume

Luckily, this is an easy process.  Simply right click on the available volume and select “Delete Volume”.


You get one more chance to change your mind.


AWS will take a short time to delete the volume:

If you get impatient (it can take a while to delete a volume and the bigger the virtual drive, the slower the delete), you can click on the Refresh button to see the current status.

Until the volume is finally gone.

Conclusion

If the AMI is not set up to delete the volume on termination, you must do so manually, or modify the AMI via the command line interface to delete upon termination.

Set up your AMIs to delete volumes on termination or remember to check and delete available volumes – remember, you are being charged to store drives you no longer use.


25 September 2010

11.1.2 zombies ate my brain, and how I got it back, part 2

There’s nowt as queer as folks

Some of you actually like doing installations. 

Well I don’t.  Perhaps that’s because I’m not particularly good at it, as anyone who has followed my blog/rants/ravings/self-flagellation know.  

As the song goes, “Every Man To His Own Profession”.  My profession (I like to pretend this is so) is implementation of Oracle EPM solutions (Essbase, Dodeca, Planning, ODI, scripting, etc.), not installing the software that makes ‘em run.  I find building a system hard enough without any added excitement up front, ta very much

There’s good news tonight

Well, the whinging can stop, because I’ve made the decision to outsource the whole mess

Angie, even though your non-reading of this blog is like having several 50-kilogram bags of delicious single estate coffee on my chest, pressing, pressing till I can’t breathe, I can’t thank you enough for your suggestion to try hosting a 11.1.2 instance.  I’ll just pop a nitro tablet and all will be well.  I hope.

My last post covered EPMCloud.com’s consultant install (everything’s there, but it’s not for production use and sits on a single server); this one will go with Full360.com’s offering.

Full360

In their own words:


We’re a global business intelligence hosting, consulting and outsourcing team. We want to help simplify the complexities of developing and managing an analytics platform for our customers. 

We'd Like to Help You:
  • Scale your Analytics Platform
  • Migrate to New Versions
  • Design insightful analytic reports
  • Develop End to End Reporting Systems
  • Reduce Consulting Expenses with remote support and outsourcing
  • Significantly Reduce Infrastructure Expenses with hosting services

It’s that last bullet point that I find so intriguing.  Time has not made me fonder of 11.1.2’s memory hogging.

The most important question and you-may-stop-reading-right-here-but-why-as-there’s-so-much-more?

How does the Full360 Oracle EPM 11.1.2 instance in the cloud work?  Flawlessly. 

So perfectly that I used a demo day on their system to take all of the notes and screen shots for a Calc Manager blog follow up to my “Why I hate (and love) Business Rules, parts 1 and 2” blog posts.  Beyond the utility of getting a new blog post on the latest release, I learned a lot about 11.1.2’s form creation and Calc Manager.  For this consultant, that is exactly why I need a cloud install of 11.1.2.

The instance was fast and rock solid.

You really could stop reading (some of you already likely have) here but for those brave souls who persevere, there’s good stuff ahead.

How is Full360’s product like EPMCloud’s?

Full360’s Oracle EPM instance in the cloud is similar to EPMCloud’s.  What does that mean?

Both company’s products share these attributes:
1.    Amazon Elastic Cloud Compute (EC2) web service for virtual hardware
2.    An EC2 Amazon Machine Image (AMI) with :
    1. 64 bit Windows 2003 Server
    2. 64 bit SQL Server
    3. All of the loveliness that is Oracle EPM 11.1.2 installed for your computing pleasure, no configuration required
3.    Use Microsoft Terminal Services to connect to the remote desktop of your EPM server in the cloud.

How does Full360 differ?

The whole Amazon EC2 experience

Full360 customers are going to have at least a basic understanding of the way Amazon EC2 works.  This is not a bad thing, as you have extraordinary flexibility with what you do with native EC2.  Remember, I think that “infrastructure” is spelt with three zeds and a couple of numbers; even I was able to handle it.  As the saying goes, if an idiot like yr. very obt. srvnt. can do it, surely you can, too.

Regardless, there is more up front work with Full360’s AMI.  This background knowledge is probably the biggest difference between Full360 and EPMCloud although both companies use the same EC2 base.

Rohit Amarnath of Full360 is also available to help navigate the morass that is Amazon Web Services even to the extent of setting up the AWS account.  I like to think of my experience as character building.

For those of you who want to follow my masochistic path, see Paul Stamatiou's guide or Amazon’s own getting started tutorial.

Here’s a brief comparison of the steps we Sandbox users would follow for either product.

EPMCloud

EPMCloud’s model is simple:
  • One AMI – 17.1 GB of RAM, 2 CPUs, 55 GB of RAM, Windows 2003 R2 Datacenter Edition, SQL Server 2008
  • Email EPMCloud when you want your instance started up
  • Get an email from EPMCloud with an server name
  • Connect to the AMI using Windows Terminal Server
  • Do work
  • Shutdown the server when done
  • Get billed based at $25 for initial start, then $5 per restart, with a $1.15/hour charge rate.

Full360

Full360’s model is more complex, but as we’ll see in a minute, more flexible:
  • Sign up for Amazon Web Services (AWS)
  • Sign up for Amazon Elastic Compute Cloud (EC2)
  • Sign up for the Full360 instances at $99/month
  • Wait for Full360 to grant access to the Full360 AMI
  • Launch your instance through the AWS console
  • Pick the Full360 Windows 2003/SQL Server 2005 Oracle EPM 11.1.2 AMI
  • Be Top Cat
  • Get billed at whatever the AMI costs per hour (see below)
  • Switch to 9.3.1 or 11.1.1.3 AMIs if you want
  • Upgrade to an-OS-not-named-Windows  (Windows isn’t the only game in town.  Real geeks use Linux.)

EC2 Instance

That flexibility is evinced when you match the Full360 AMI against whatever 64 bit Windows and (Full360’s AMI uses SQL Server Express, so it’s just plain ol’ Windows) EC2 instance you want. 

Btw, I absolutely love these names – marketing genius.  No boring “regular,” or “normal,” or even “big” to be seen.  Perhaps the yet-to-be-released 128 GB 16 way box will be called “Nathan Explosion ?

Instance name
RAM in GB
CPUs
Disk in GB
Cost/hour
Extra Large Instance
15
4 virtual cores with 2 EC2 Compute Units each
1,690
96¢

High-Memory Extra Large
17.1
2 virtual cores with 3.25 EC2 Compute Units each
420
62¢

High-Memory Quadruple Extra Large Instance
68.4
8 virtual cores with 3.25 EC2 Compute Units each
1,690
$2.48

While the AMI would likely either be the Extra Large Instance or High-Memory Extra Large Instance for your consultant (train yourself, prototype, have a place to answer all of those OTN questions you obsessively go after) instance, surely you will at least once pair up the Full360 11.1.2 AMI up against the High-Memory Quadruple Extra Large Instance.  That last one sounds like the aft turret on HMS King George V.  Talk about not needing a bigger boat

What’s installed?

This is an Amazon EC2 Windows 2003 Server/SQL Server Express 2005 instance.  The rest is Full360’s install of the 11.1.2 stack.

Install
Config
Windows 2003 Server
Yes
Yes
Foundation Services
Yes
Yes
Workspace
Yes
Yes
Essbase
Yes
Yes
EPMA
Yes
Yes
Calc Manager
Yes
Yes
EAS
Yes
Yes
HPS
Yes
Yes
Planning
Yes
Yes
HFM
Yes
Work in progress
Financial Reports
Yes
Work in progress
Web Analysis
Yes
Yes
ODI
No
No
FDM
No
No
MS Office 2007
Yes
You must supply a valid key
SQL Server 2005
Yes
Part of the AMI
GhostScript
Yes
Yes
Essbase Studio
Yes
Yes
Smart View
Yes
Yes
Excel add-in
Yes
Yes

Remember, you are the Windows administrator on the AMI.  Need something you don’t see above?  Install away.

Logging on

We’re talking Windows and EC2 – it’s the same as EPMCould, so Windows Remote Desktop Connection to the server (I cleverly forgot to take a screenshot but I think we’re all familiar with this dialog box).

Ever so slightly scary

Btw, do not be alarmed by this rather ominous looking dialog box.  I got the same thing from EPMCloud:

See this EC2 security guide to calm your madly beating hearts.

Mundane?  Not a bit.

Just click on the Yes button and off we go to the unexciting-and-yet-exciting-all-at-the-same-time Windows desktop:

That shade of blue has been my boon companion since Windows NT 3.5.1 (geezers will know why that mini-VAX of an OS was far, far better than NT 4.0 and its descendants).  So not so much excitement, you might think, but that’s where you’re wrong.

The Start Menu

And this is why the ordinary isn’t ordinary.

I had not a single blessed thing to do with this.  Thankfully.

What kind of server?

Don’t be alarmed by the 7.5 GB of RAM – that’s just the EC2 instance Full360’s AMI ran on.  It could have just have easily said 17.1 GB.

Client Connections

As already noted, Microsoft Office 2007 is include on the server, so for Sandbox use, there’s really no need to connect in a client like Excel.  NB – You must have a valid Office key for this to work.

Performance

Despite being connected to a 7.5 GB server (well under the completely unofficial 12 GB threshold for 11.1.2) performance was more than adequate.

Wait, you say (you are paying attention, aren’t you?), what about your 8 GB laptop that couldn’t run 11.1.2 worth a tinker’s damn?  I have a feeling that Amazon’s hard drives/SAN/who knows what are faster than my 7,200 rpm laptop drives.  Or it could be that I can’t configure an 11.1.2 installation to save my life.  I don’t know which explanation I prefer.

I am semi-afraid to hook up the AMI to the High-Memory Quadruple Extra Large instance as I may never want to go back to the less than ½ price High-Memory Extra Large instance.

Price

Let me just get this out in the beginning -- Full360’s overall monthly price in the below usage model is cheaper than EPMCloud’s. 

Having said that, the product that is right for you and cheaper is going to depend on:
1)    How often you use it.  You must first go through $99 of EPMCloud’s on-demand usage per month to see savings from Full360.  In my usage model below, you could use the EPMCloud product till 14 September with 43 hours of server time for that amount of money. The equivalent usage pattern with Full360 would cost approximately $126.  Alternatively, by the end of the month Full360’s product is cheaper.
2)    How flexible do you need your 11.1.2 instance to be.  If you’re going to clone it, snapshot it, put it against different EC2 server instances, Full360 has more options because more of EC2 is exposed.
3)    How involved in the world of Amazon EC2 you want to be.  As I wrote, it isn’t hard, but it made me think a bit.  Geeks like this kind of challenge, sometimes, but often just working on the stack is what’s important.  Real life story -- I know of one consultant who recently emailed me after reading the EPMCloud post – I told him about Full360’s EC2 requirements and despite the lower price (again, assuming the usage model below), he went for EPMCloud because it was easier for him.

The above are not trivial considerations.  Think long and hard about how you are likely to use the products and then drop Full360 and EPMCloud an email and decide for yourself.

So how much does Full360 cost?

1)    There’s a $99/month AMI fee (see the ODTUG connection below).  That’s the ability to access Full360’s fully tested out Windows 2003 Oracle 11.1.2 EPM instance.
2)    There’s an hourly cost based on Amazon’s EC2 pricing.  If you went for the High-Memory Extra Large instance, the cost is 62¢/hour.

The ODTUG discount

You are a member of ODTUG, yes?  And you’re voting for the board of directors, yes?  And you’re voting for me, yes?  I will now go beyond shameless and highly recommend Tim Tow and Angie Wilcox as candidates.  If you don’t vote for me, at least vote for Angie and Tim.  Remember you have till 2 November 2010. 

Beyond the many existing reasons to be a member of ODTUG (Kaleidoscope, the best Hyperion conference ever, the beyond-invaluable Technical Resources, the technical journal) there’s yet another reason to join – Full360’s ODTUG discounts.  The monthly charge is half price for ODTUG members.  Instead of the normal $199/month, the price is instead $99.  The startup fee is $1 instead of $99.  Good grief, that’s a first year savings of $1,298.  Does that cover your $175 yearly ODTUG individual membership?

I am going to assume that you too can do the math, hence I will use the ODTUG special prices.

Cameron the consultant’s use case

Before we dive into that cost analysis, let’s again review how an 11.1.2 instance in the cloud is likely to be used by a mythical consultant named “CL” (who is this guy, anyway?).

My usage model will likely be broken into two types:
1)    Common, everyday
    1. An hour or two before work and at night (yes, I am sad)
    2. A few hours every weekend
2)    Extraordinary, short bursts
    1. Learning new stuff for a project, e.g., Calc Manager the right way
    2. Proof of concepts/prototypes


Startup
Hours
Cost @
62¢/hour
Why?
1-Sep
$99.00



2-Sep




3-Sep




4-Sep

4
$2.48
Tire kicking
5-Sep




6-Sep

2
$1.24
Training
7-Sep




8-Sep

1
$0.62
Training
9-Sep

8
$4.96
Prototype
10-Sep

8
$4.96
Prototype
11-Sep

4
$2.48
Prototype
12-Sep




13-Sep

8
$4.96
Demo to client
14-Sep

8
$4.96
Tweak prototype
15-Sep

2
$1.24
Demo to client
16-Sep




17-Sep




18-Sep

3
$1.86
Blog writing
19-Sep




20-Sep




21-Sep




22-Sep




23-Sep




24-Sep




25-Sep




26-Sep




27-Sep

8
$4.96
Prototype
28-Sep

8
$4.96
Prototype
29-Sep




30-Sep





$    99.00
        64.00
$39.68









$138.68


If this usage model even approaches reality for me, it would take me almost two years to make this as expensive as that IBM W510 or four years to match the cost of the Dell Precision 4500 that I lamented over in my last cloud post.

I think Full360 is cheaper than the hardware route, don’t you?  Remember, if 11.1.3 or 11.2.1 or whatever the next big release is (I have zero insight to this stuff so who knows what it will be called) and it now requires 20 GB of RAM to run, just move on up the EC2 instance scale.  Try that with a laptop.

Finally, I’ll be able to buy some sexy looking thin laptop next time around (my current brick weighs about seven pounds and reminds me of an aircraft carrier).  Maybe my back will stop hurting when I travel.  My wallet surely will.

Conclusion

So there you have it, yet again.  A complete 11.1.2 installation, tailor-made for consultants or clients who want/need to kick 11.1.2 tires.

It’s a shame that 11.1.2 is from beyond the dead, intent on nothing more than eating our computer’s brains.

But it’s fantastic that there are two great companies offering fully-vetted 11.1.2 instances that we can start learning from and building on.

I have to say that I never really understood the appeal of cloud computing before these two posts – now that I get the flexibility, power, and affordability, I’m likely boring everyone I meet with The Way.  You really, really owe it to yourself to check it out.

Now I can finally get back to hacking Essbase, instead of figuring out how to make it run.