The respondents to Computerworld’s annual forecast survey voted .Net as a hot programming skill for ‘09. It is surprising there is no mention of Java/JEE in the top nine list of hot skills. Per the survey, the top skills for ’09 are:
1. SAP, .Net in application development category
2. Help desk/technical support
3. Project management
4. Networking
5. Business Intelligence
6. Security
7. Web 2.0
8. Data center
9. Telecom
The complete article is available here.
The current economic crisis has resulted in massive layoffs across the world. LiveMint.com, a subsidiary of Wall Street Journal, reports there is at least one layoff every ten seconds. In the US, for the month of November alone, there have been more than 500,000 jobs slashed. Techcrunch layoff tracker reports over a hundred thousand job cuts in tech companies. However there is some positive news from various job sites around the web.
Mentioned below are some resources from around the web to survive a layoff.
What is the value proposition of SaaS integration?
Software as a Service (SaaS) Integrators integrate an enterprise service with Software as a Service provider or integrate two or more SaaS providers. The SaaS based integration is billed based on consumption unlike a typical enterprise in-house integration which follow chargeback models. The value proposition of SaaS integration is in reduced complexity in accessing information, cost savings and increased reusability of the data. SaaS integrators focus on integration through configuration rather than, coding.
Who are the players?
CastIron, XAWare, Boomi, Jitterbit and Hubspan are the players in this space. From their product white papers and features, all of them seem to have some similar capabilities such as:
1. Support for XML data formats
2. Leverage visual integration pattern
3. Mediation services
4. Data transformation services
However the delivery format from CastIron is through a hardware appliance and is different from other vendors in this space.
More analysis to come later.
Per a recent posting on programmable web, the number of Web API’s has crossed the one thousand mark, an important milestone in the web world. The example posts mentioned below support the increasing importance of having a Web API.
1. Sixty percent of eBay product listings come from their API
2. Twitter’s API processes 10 times more traffic than their website
3. Amazon Web Services consume more bandwidth than the entire global network of Amazon websites
The value add for an enterprise having a web API provides thousands of points of presence rather than, one point of presence through a website. The multiple points of presence lead to creation of new products and services increasing the possibility of revenue making channels.
Hosting a web API requires configuration and setting up components for analytics, caching, security, credential management and integration with the enterprise repositories among others. However, such services are not core functions to the business of many enterprises. Sonoa offers features in addressing such needs in multiple deployment modes such as hardware appliance, virtual appliance and SaaS in the near future.
Sonoa was founded in 2004 and the first product was shipped in Q2 of 2007. The management team comprises of members with a background in middleware, applications and networking. Sonoa closed $10 million in series C round of funding led by Third Point Ventures, in October of this year. The other investors in the company are Norwest Venture Partners, Bay Partners and SAP Ventures, a division of SAP AG.
Sonoa is currently distributed as hardware and virtual appliances. The virtual appliance is currently available for VMware only, however, Sonoa plans to make it available in other deployment formats such as virtual appliance deployable in Amazon cloud and SaaS offering.
Enterprises who are concerned about performance degradation when accessing their API’s due to a spike in the number of consumers, would find Sonoa appliance interesting. Sonoa offers XML traffic acceleration because of built in TCP optimization thereby offering optimal speed for data transfer between the Web Service endpoints and the gateway (i.e. Sonoa appliance). The appliance can also be configured in a cluster for high traffic needs, high availability and failover.
[Click on the image for an enlarged version]
Sonoa also offers features such as: analytics component providing visibility when servicing consumers, a messaging engine for content based routing and hardware acceleration for XML processing. A policy engine for execution of SLA’s providing control when servicing consumers. Mediation capabilities such as: a) Protocol Mediation, for example HTTP(S) to JMS/TCP-IP b) Credential Mediation, for example SOAP/WS-Security to SAML c) Payload Mediation, make integrating Sonoa in an enterprise cost effective.
On the security front, Sonoa offers data transfer over encrypted channels and credential mapping. Sonoa extends through Eclipse based design studio for creating complex policies.
The Web API economy is bound to grow and there would be more enterprises and startup companies who would open up their functionality through API’s. Alternate API infrastructure management solutions in the cloud (SaaS) such as Mashery and 3Scale are available however for enterprises who believe in the strategic value of the data resulting from the consumption of their API, would find an in-house solution such as Sonoa compelling.
The value propositions of monitoring a Websphere MQ environment (or the Queue Manager network) are mentioned below:
1. Provide business service visibility i.e. how many cross sell offers (messages) were processed?
2. Precise measurement of the Queue Manager throughput for capacity and performance planning and architecture fine tuning.
3. Helps in troubleshooting issues related to Queue Manager.
This posting discusses three tools that can monitor a Websphere MQ environment. Unlike Java application performance management tools, these tools try to be different from each other.
1. QPasa from MQSoftware is arguably the market leader in this segment and probably the first MQ monitoring solution. QPasa supports monitoring metrics of two kinds, real time and historical. QPasa is based on “publish and subscribe” architecture, thereby enabling parallel processing of performance messages. This architecture makes QPasa an appropriate choice for monitoring high volume MQ environments. QPasa offers rich features such as custom dashboard reporting providing business visibility into messages processed and configuration management. However, it requires agents to be deployed on the servers hosting the queue manager in order to instrument them. On the downside, QPasa requires a high learning curve.
2. QFlex from Netflix uses a simple architecture of near real time metrics collection based on Websphere MQ’s reset statistics. QFlex is agent-less requiring no configuration or deployment overhead. One can start using the tool in less than an hour. QFlex offers features such as change management, queue browse utility that can be used to move messages from one queue to the other or in maintaining poison messages. Netflexity provides excellent technical support. However, QFlex does not offer report customization.
3. Statwatch originally from Reconda Software, was later acquired by MQSoftware, provides historical monitoring of message traffic, message tracking thereby enabling measurement of time taken for message traversal across queues. Statwatch relies on Websphere MQ API exits that are deployed as agents on the queue managers that need to be monitored.
If you have used any other product please share your thoughts/experience in the comments.
A load test, when conducted before a web site goes live, helps in detecting and responding to scalability and performance problems much before the user is aware of such issues. Executing a load test from the cloud offers the following benefits:
1. Zero administration costs that would have otherwise incurred due to setting up an infrastructure for performance monitoring and load test generation.
2. A sub optimal performance infrastructure skews the results of a load test. However, when using cloud test tools, such issues don’t arise.
3. Measure the perceived end user’s response time of the application across DSL, cable, dial up, 3G and T1/T3 connections.
The Tools
SOASTA CloudTest provides load testing as a service. CloudTest is available in two forms a) as a service and b) as an appliance. Both these offerings can be used to test an application hosted in an enterprise behind a firewall or a web application hosted in a cloud (such as Amazon EC2).
Monitoring is done in two modes, using an agent or in agent-less mode. In agent mode, an installation of the agent (called a conductor in CloudTest Terminology) on the target server is required which would collect the data and transfer to the CloudTest Appliance or Service over HTTPS connection. While CloudTest enables easy creation and execution of load tests, it also provides visually appealing reports of the test results. The test results can be searched by message or error or by value, simplifying the post mortem analysis on performance bottlenecks. If your web application is hosted in San Francisco area how do you know what is the response time perceived by a user accessing in Beijing, China? CloudTest does not help in such situations. Here’s where testing tools from Keynote Systems help.
Keynote Systems has various products that are offered as a Service, from generation of tests, monitoring, diagnosing transactions to reliable alerting. Keynote Systems provides visually appealing reports, easy creation and tear down of load tests. However the value proposition is from their unique offerings,
1. The ability to execute a load test on a web application from a location in their global network which, includes approximately 240 cities across the globe with more than 2400 computers, obviously such a test would paint a real picture of the end user’s experience across various geographic locations.
2. Combine the above with the ability to measure application response time across various connection channels such DSL, 3G and T1/T3 among others, provides a comprehensive coverage of the end user’s experience across various channels.
Keynote KITE offers a free version that will allow you to execute tests from five different cities. However, testing from more locations would involve licensing.
Other articles of possible interest:
The Platform as a Service (PaaS) landscape
What does Oracle cloud offering mean to the enterprise
A list of on demand message queue (MQ) providers
ServiceNow, a SaaS offering of ITIL
VMware and Citrix want to be enterprise cloud enablers
Top five Java application performance management tools
The ever so interesting Tim O’Reilly does an interesting analysis on the future of cloud computing (as understood in the context of IaaS), whether it would evolve into an outsize profit business as envisioned by Hugh Mcleod in his post titled “The cloud’s best kept secret”. Some interesting points from the post are mentioned below:
1. Presents an interesting discussion on three kinds of cloud computing models. Summing up, they are:
a.Utility Computing (aka IaaS): Providing computation power that is billed based on consumption. Examples: Amazon EC2, GoGrid and AppNexus among others. The bottom line that these services are targeted to a developer audience rather than end users limits the network benefits of data. Thus IaaS would be a commodity business.
b.Platform as a Service (PaaS): PaaS is a hosted application stack enabling applications to be built and deployed over it. In other words, PaaS provides a much higher level abstraction of an infrastructure than IaaS. Examples in this category include Google App Engine, Mosso, open source offering 3Gen and Salesforce’s Force.com among others. However Tim asks an interesting question of what is value add to developers in one of these platforms from other developers on the same platform? None except Salesforce.com which seems to provide some benefits to developers.
c.Cloud based end user applications (aka Cloud Services): Any web application is a cloud application. For example Google, twitter, Facebook, Amazon etc. This type of cloud computing has network effects and has the potential of growing into an outsize profit business. Salesforce’s force.com platform is in alignment with this model through the developer ecosystems that it provides.
2.The Law of Conservation of Attractive Profits
“When attractive profits disappear at one stage in the value chain because a product becomes modular and commoditized, the opportunity to earn attractive profits with proprietary products will usually emerge at an adjacent stage.”
Example: When IBM began building its PC from off the shelf parts it drained value from the hardware business turning it into a commodity (low margin) business. However the profits didn’t go away, instead, they migrated to software i.e. from IBM to Microsoft.
3. Applies the above mentioned law to Larry Ellison’s point, that cloud computing is not a profitable business model, and argues that Larry “is dangerously wrong for the strategic future of Oracle”. Why? Because “it’s not the database software that matters, but the data that it holds, and the services that can be built against the data”.
The bottom line is that neither would IaaS or SaaS alone can grow into a huge business, however, PaaS can.
Other articles of possible interest:
1. The Platform as a Service (PaaS) landscape
2. What does Oracle cloud offering mean to the enterprise
3. A list of on demand message queue (MQ) providers
4. ServiceNow, a SaaS offering of ITIL
5. VMware and Citrix want to be enterprise cloud enablers
© 2006-2008 Byteonic.com
All Rights Reserved
Content is presented without any warranties or guarantees. Opinions expressed are based on the analysis done at the time and are subject to change without any notice.
