Sunday, December 18, 2011

ITE 221 - Fall 2011 - Chapter 13


Article Review “Moving to the cloud in 2012? Look out for these pitfalls”
By Nancy Gohring, Dec 16, 2011

In this article industry experts discuss some of the most common mistakes when moving to the cloud.
Failure to set up redundancies for disaster scenarios. John Engates, CTO at Rackspace, pointed out that developers, "often assume that the cloud service takes care of security, application availability, backup and recovery, and ensuring service performance."
Not planning for data center failures. Many cloud users don’t design their systems to ensure they are load balanced and can shift workloads to different zones in the event of an outage.
Failing to accommodate for potential bandwidth issues. Joshua McKenty, CEO of Piston Cloud, noted, "We will continue to see problematic issues with public cloud bandwidth limitations.” Many companies simply switch to private clouds rather than search for other options.
Vendor Lock-in – Engates hoptes to see “more progress toward a de facto standard.” In the coming year, HP, Dell and Rackspace plan to launch their OpenStack-based cloud services for this purpose. The goal of OpenStack is to Open Source Cloud Computing platform that will meet the needs of public and private clouds regardless of size, by being simple to implement .

ITE 221 - Fall 2011 - Chapter 12


Article Review “RAID – Storage Made Smart”
Written by Harry Fairhead, 23 Nov 2011

This article provides some very good rationale for utilizing Redundant Arrays of Inexpensive Disks (RAID). It describes RAID as connecting multiple drive units together to make them function as a single “virtual” storage device. This allows you to treat the RAID system as if it was a single disk drive, although in most cases, it consists of multiple physical drives.

The article provides three key reasons to use RAID:
·    Reliability - As an electromechanical device with moving parts, a hard drive will eventually wear out. By creating redundancy through the multiple drives, you reduce dependence on a single device, which improves reliability of the overall storage.
·    Efficiency - A RAID system can be made to appear faster than any of the single drives it consists of by adding the total transfer of the two or more disks the total rate should be higher.
·    Scalability - Although you can increase the storage available simply by adding additional drives without the need to tie them together into a single virtual drive, users prefer to see a single storage volume that contains all their data.

Thursday, December 15, 2011

ITE 221 – Fall 2011 - Chapter 11

Article Review “What’s next with hypervisors?”
By John Dix

This article discusses upcoming trends in hypervisors with a panel of experts to include Al Gillen, an analyst IDC who tracks virtualization developments, Kerry Kim, director of solutions marketing at SUSE, and Adam Jollans, program director of IBM's Linux and Open Virtualization Strategy.
A hypervisor, also called a virtual machine manager, is a program that allows multiple operating systems to share a single hardware host. The hypervisors controls the host processer and resources so that each operating system appears to have the resources individually.
One major issue discussed is the differences between proprietary and the open source hypervisors. Currently the pace of innovation is much faster with open source technologies. In addition, open source development allows for more service and support options. However, this rapid development isn’t always enticing to corporate information officers. Many are overwhelmed or even intimidated by it. As Kim puts it, “It's not because customers don't want innovation; they just can't really embrace the innovation as quickly as it comes.” Instead, they tend to stick with established relationships. Another benefit of sticking with the same vendor of their other components is that they will have a single stack of software.
Another concern with hypervisors is how they will interact with a cloud provider that has implemented a different type of hypervisor. Do to this concern most companies “support customer choice” by ensuring compatibility with “various third-party tools, whether they're open source or proprietary.” Many companies are leaning towards VMware because it has a reputation as having less of a platform agenda.

Sunday, December 4, 2011

ITE 221 – Fall 2011 - Chapter 10


Article Review “Five Key Enterprise Development Trends”
By Daryl Taft posted on eWeek.com on 12/04/2011

This article discusses five key enterprise development trends including HTML5; the prevalence of Web, mobile and cloud development; advances in languages and integrated development environments (IDEs); the maturation of the Java PaaS; and the use of open-source in Analytics Agile Application Lifecycle Management (ALM).

HTML5 is a key consideration for developers. According to Al Hilwa, an analyst with IDC, “By 2013, we will reach a point where 90 percent of smartphones and tablets will sport HTML5-capable browsers.” Microsoft has already adopted HTML5 as the standard for Windows 8 and Internet Explorer 9. Adobe has stopped its development projects to support Flash technology for mobile browsers and Microsoft may even do the same with Silverlight, its development framework for building Web and mobile applications.
The next trend for developers to be aware of are advances in languages and integrated development environments (IDEs). Current
Web-based development environments like Eclipse Orion, Cloud9 IDE, and eXo Cloud IDE are already becoming more popular. These make the browser in the mobile device the IDE.
The third major trend is referred to as the “big-data and analytics craze.” This is referring to the need to capture and analyze the plethora of coming from intelligent devices and social media. As a result, IDC predicts the market for intelligent systems will soon represent a $520 billion industry.
Next is the maturation of the Java platform as a service (PaaS). Oracle’s Java Cloud represents the culminates of PaaS offerings and sends tells that Java developers to start looking at these solutions for deployment of new applications.
The final trend noted is Agile Application Lifecycle Management, which is the continuous process of managing the life of an application through governance, development and maintenance. More and more developers are integrating new tools to support Agile development. Mik Kersten, CEO of Tasktop Technologies referred to it as the “open-source powered tidal wave changing how developers work and collaborate”.  

Saturday, December 3, 2011

ITE 221 - Fall 2011 - Chapter 9


Article Review “It’s Time to Get our 10 Gig On: A spate of new products is putting high-performance 10-Gbps Ethernet within reach of almost every organization that wants it. “

By Art Wittmann posted on InformationWeek November 16, 2011

This article discusses new products that are available to allow companies to switch from 1 Gbps to 10 Gbps Ethernet. They advances are mainly due to advances in merchant silicon that have improved density. The good thing about this is that is significantly reduces the price per port making now a good time for companies to upgrade.

It specifically identifies the following new technology products:

  • IBM RackSwitch G8316 - 40-Gbps, 16-port aggregation switch in a 1U form factor. According to IBM if you combine it with the G8264 at the access layer, it can accommodate up to 64 1/10-Gbps SFP+ ports, therefore this is a good choice for density.
  • Cisco Nexus new line cards - 48 1/10-Gbps SFP+ ports, which provide a very flat data center network with up to 768 ports in a 7018 chassis.
  • Arista 7050 - fixed-configuration switch with 48 1/10-Gbps ports and four 40-Gbps uplinks. It’s even more economical than Cisco, but Arista is a newer company.  


The article also notes that there are many other vendors also releasing upgraded data center and campus networking products with both 10- and 40-Gbps Ethernet ports at much higher density.