Comments for Neosavvy, Inc

Comment on I moved…out of the flippin cloud by Adam Parrish

Adam Parrish — Thu, 19 Jan 2012 02:54:54 +0000

Daniel:

You ask an excellent question and for the essence of discussion I will outline how I came to the purchase decision and my budget.

Firstly summing the cost of services I paid for last year – that are now replaced by purchasing the server
3334/yr hosting (just to rackspace)
864/yr harvest (depending on user count but this is what quickbooks says i spent last year)
1750/yr atlassian jira studio (i just checked quickbooks and misquoted the amount in the blog post)

Total yearly spend: $5948

Now keep in mind the spec of the hosting was 3 boxes (CentOS with 80GB HDD and dual core processors with 2GB memory)

Server Purchase:
1U SuperMicro 24 Core Intel Xeon with 72GB RAM and 8x256GB Crucial SSD Hard drives and hardware raid controller and 10GBe card
2U QNAP NAS with 8x 2TB Seagate Constellation 3 Enterprise Harddrives
10GBe expansion for QNAP to create 10gbe connection to server
12U Rack unit to house these units and future server expansion

Total cost $16,330 – amortized across 3 years $5443/yr or assuming I get 5 years of service out of this equipment $3266/yr.

Now just by pulling my services out of the cloud I will end up saving $500/yr over 3 years. But also realize I have MUCH more capacity than I was getting and can spend money on licenses for hosting the software myself in the future and also have room to write my own services and expand my internal capabilities.

Hope this helps you look deeper into my move out of the cloud, and maybe it’ll rationalize a similar move for you if you are starting or running a small business yourself.

Comment on I moved…out of the flippin cloud by Daniel

Daniel — Wed, 18 Jan 2012 01:20:09 +0000

“Actually purchasing hardware and leveraging CentOS with XenServer Hypervisor has enabled me to spin up 10 beefy virtual servers and there is room for 10 more for additional processing power.”

But what did it cost for that hardware and hosting?
Basically what is the cost/benefit comparison between rackspace’s cloud and “real” hardware.

Not that I’m arguing for/against the cloud mind you, everything has its place and the cloud isn’t the solution for everything, just wondering.

Comment on Applying the Scientific Process to test runtime performance of HGroup vs Group by Adam Parrish

Adam Parrish — Tue, 23 Aug 2011 20:43:51 +0000

Paul – I restarted between each session and then in a non-recorded session I actually ran them sequentially.

I have also done the same code with “addElement()” calls after it instantiates the object and have seen some surprising results. HGroup actually outperformed Group in some instances of the runs.

Comment on Applying the Scientific Process to test runtime performance of HGroup vs Group by Paul Taylor

Paul Taylor — Mon, 22 Aug 2011 17:51:27 +0000

Did you run each tests in the same session, or did you restart the app each time you ran the test? There’s a nearly 300% difference between the deltas in your first and last test run. If you ran the tests in the same session, I suspect the Flash Player needed to allocate another block of memory for the new instances. In that case, you’re not just measuring instantiation time, and the results are erroneous.

Regardless, the results show this is a silly argument to have. Even with the possible memory allocation, there’s roughly 0.06% difference between instantiation times. And since item renderers are virtualized, you’ll probably never create 10,000 of them at the same time anyway.

Use the Group with the API that reduces the amount of code you write, forget about worrying how many super classes it has. If it really mattered we’d all be using alternative component frameworks anyway.

Comment on Applying the Scientific Process to test runtime performance of HGroup vs Group by MR

MR — Mon, 22 Aug 2011 12:37:41 +0000

Have you looked at the measurement/layout code for Group/HGroup yet? I’ve not spent (any recent) much time with Flex 4, but I do recall at MAX a couple years ago Adobe saying that they made several improvements and optimizations in this sense. Maybe the performance impact in Flex 4 isn’t as great as Flex 3.

Comment on Building a Simple Login Mechanism with Flex, Spring-Flex, and Spring-Security by krish

krish — Tue, 02 Aug 2011 18:29:56 +0000

Absolutely amazing. I am new to IntelliJ env. could you please help me in getting this sample working.

Comment on Building a Simple Login Mechanism with Flex, Spring-Flex, and Spring-Security by badtnc

badtnc — Fri, 01 Jul 2011 18:35:17 +0000

can you please let me know how to run this application

Comment on Building a Simple Login Mechanism with Flex, Spring-Flex, and Spring-Security by VELO

VELO — Wed, 15 Jun 2011 23:39:53 +0000

I never cease to get amazed with the gratitude people have with those that do volunteer work…. truly amazing!

Comment on Building a Simple Login Mechanism with Flex, Spring-Flex, and Spring-Security by Adam Parrish

Adam Parrish — Wed, 15 Jun 2011 20:11:08 +0000

Hey:

Sorry it took you 3 days to run the example. I don’t constantly monitor my posts to make sure they still run as this one is easily 2 years old now.

I try to help people out so please don’t hesitate to email me or post questions as I am happy to help you give it a shot and help you if you are having trouble.

I easily spent a few weeks creating this example in my free time so I hope it is helpful to others as it was quite a lot of work to put together for me.

-Adam

Comment on Building a Simple Login Mechanism with Flex, Spring-Flex, and Spring-Security by fofo

fofo — Thu, 09 Jun 2011 17:25:03 +0000

thanks for the contripution, but you could add a toturial to help people get this running fast and not to let every body wast 3 days to run the example!!!!!