With the Web accounting for an increasingly large piece of the retail pie, having Web applications that are fast, intelligently designed, and able to scale is important. After all, the customer is only one mouse click away from abandoning you - and perhaps forever. Freaking retailers out with such thoughts is the purpose of a …
Cool that HP is making money from folks that think that *aas is a long-term valid thingie.
I'm only sad that I made the mistake of missing this particular short-term niche money maker.
So, what is new?
Mercury Interactive used to do that in the late 90s. There are piles of examples of retail/financial customers that lost millions during eCommerce high time (xmas). Mercury whent heavily into SaaS around 2002-6 with great success as benefits are achieved much faster by leapfrogging implementation.
Can someone tell me what HP has brought as "new approach" to the table before I fall asleep, or get frustrated by their lack of innovation??
Regards.... ex-head of performance testing at Mercury.
Except that many deployments cant easily be "given to HP to run". 100 different services, DBs, exotic networking environments, latency simulation for global distribution...
Sounds perfect for some crappy shopping site, but useless for most grown up websites comprising something other than a buggy LAMP stack.
I know where I work this simply wouldnt work and we've considered slapping down a credit card on amazon ec2 and firing up 10,000 VMs to ddos our application.
re: Old school
I understand how at first glance how this may seem like a retread of Mercury's (and now HP's) extremely successful SaaS Performance Testing solution. I've been a user and product manager for the performance SaaS service for 9 years.
HP Elastic Test brings to market 3 new innovations:
1. Ability to scale to huge loads (e.g., hundreds of thousand virtual users) in an on-demand fashion (i.e., minutes)
2. New utility-based pricing and built-in usage metering, so it's extremely flexible in pricing
3. Click-n-script technology, especially for Weg 2.0 website, which greatly speeds up the testing cycles (during tight timeframes) and provides more accurate performance measurements (all asynchronous communication is emulated)
Thanks for reading!
HP can't even stress-test their OWN software properly. I'd find it hard to believe the results of any stress test they did.
By stress-test I mean:
Lets install HP software. Reboot system. System goes into infinite loop because of syntax error within code. Blame sys admin saying "you must have installed it wrong"! Though quite how you can get a one-line install command with no options wrong sort of escapes me (rather like HP's grip on reality).
Sounds like a bad idea - should you really be capacity planning based on HPs set-up and load test in a virtual environment? If its just to see where bottlenecks are then its not much use either as you will still need to load test in your own environment to find any specific set-up issues with your infrastructure.
Unless you actually plan to run the app in HPs elastic data-centre or whatever then its not a particularly useful service.
- Vid Hubble 'scope scans 200,000-ton chunky crumble conundrum
- Bugger the jetpack, where's my 21st-century Psion?
- Google offers up its own Googlers in cloud channel chumship trawl
- Interview Global Warming IS REAL, argues sceptic mathematician - it just isn't THERMAGEDDON
- Apple to grieving sons: NO, you cannot have access to your dead mum's iPad