Week Nine

Week Nine

This week I decided to focus my research on Software as a Service (Saas) Performance Metrics, focusing specifcally on synthetic transactions.

When monitoring a SaaS application as a consumer you focus on two questions:

1. Can I access the service?

2. Does the service perform as expected?

The company I work for has a way to check each website for performance. Our textbook labels this as a "synthetic transaction". Our company has a generic user ID that checks a website and logs in to check functionality. If there are issues, it alerts to our command center. If it is successful it waits two minutes and tries the check again. Overall, we rely on synthetic transactions to ensure our websites are function and end-users are able to view important websites at all times.



Another company I found, CA Technologies, uses synthetic transations at more than 90 monitoring stations in over 40 countries. Their monitor replicates real-user transactions that help identify issues. They claim that this monitor will also help resolve problems before they affect end users. They use this monitor to optimize the end-user experience. By using synthetic transactions it provides an up to date insight on performance.

I think these monitors are very important, especially for a company that works with customers. Any impact to service could result in a loss of business, and synthetic transactions help find the issue before it affects the bottom line.

Source:
http://www.ca.com/us/opscenter/ca-application-performance-management-cloud-monitor.aspx

Week Eight

Week Eight
This week I decided to dive into incident management. I chose this topic because the company I work for takes this topic very seriously. Every minute operations are impacted means money lost. According to the ITIL an 'Incident' is any event which is not part of the standard operation of the service and which causes, or may cause, an interruption or a reduction of the quality of the service. This is how my workplace also defines an incident.

The objective of Incident Management is to restore normal operations as quickly as possible with the least possible impact on either the business or the user, at a cost-effective price. For example, at my workplace, every server has a backup server. When the primary fails, we are able to move operations to a backup server. Cost-effective? Perhaps not. But it minimizes impact on the business, and the impact does not reach customers. When operations are moved off of the problem server it is able to be investigated and repaired with as much time as necessary. This reduces room for error and allows for an in-depth "what happened" analysis to be completed.

The chart below is the way that my work handles incident type situations.




1. Triage - contact the correct application teams and notify them of the problem. Start a phone bridge to have application support all in one place.
2. Analysis - What is the issue? What caused the issue? Who is impacted?
3. Reaction - Do we fix the issue? Can we restore the application? Should we switch from primary to secondary locations?
4. Restore - Resolve the issue by the solution determined in step 3.
5. Post-Mortem - What happened? What can we have done differently? What can we do to prevent this from happening again?

Sources:
http://www.itlibrary.org/?page=Incident_Management
http://www.cisco.com/web/about/security/intelligence/worm-mitigation-whitepaper.html

Week Seven

Week Seven
This week I decided to dive in to some of the pricing options of migrating to the cloud. As I already wrote an assignment on Amazon EC2 I decided to focus on Rackspace and Windows Azure. Some providers offer pay-as-you-go plans, monthly/yearly plans, or a combination. For the purposes of this blog I am focusing on Rackspace's Performance 1 plan and Window Azure's pay as you go plan.

Rackspace
Performance 1 plan - Best suited for web servers, batch processing, network appliances, small databases, and most general-purpose computing workloads. Rackspace has a minimum charge of $50 per month.

Benefits:

• High-performance, RAID 10-protected SSD storage
• Redundant 10-Gigabit networking
• From 1–8 vCPUs

Windows Azure
Pricing is based on the size of the VM, number of VMs, and number of users. To help companies calculate their estimated monthly cost, Azure offers a full pricing calculator. 


 Pay as you go benefits:

• Zero upfront, cancel anytime
• No long term commitment  

6-month monthly pay (starting at $500/month) benefits:

• Up to 27% savings vs. Pay as You Go plan 

Week Six

Week Six

This week I decided to continue with the topic of our homework, Service-Level Agreements (SLA). According to our textbook a service-level agreement outlines the expected level of service a customer can expect to receive from a service provider, the metrics used to measure said service, and the roles and responsibilities of both the service provider and the customer. It serves as an intermediary between the cloud service provider and the client organization.



According to an article by Wired, there is confusion about the importance of SLAs. In addition to setting expectations for companies, an SLA also acts as a guide for handling potential problems. Ironically the article addresses an outage by Amazon EC2 (which I discussed in my homework assignment). Amazon had an outage in April 2011 that was almost four days long, but because of the wording in the SLA, their agreements were not technically breached.

Wired's Lessons that have been learned with SLA's:
1. Read your cloud provider’s SLA very carefully
2. Get technical staff involved to validate SLAs against common outage scenarios
3. Have contingency plans in place to support worse case scenarios

The bottom line is that an SLA is necessary if you're going to do business in the cloud. I really liked how Wired broke down the importance of an SLA when they said "Most people require a blueprint for architects and contractors to start building a new home and similarly would expect a new car to come with a warranty. An SLA serves as both the blueprint and warranty for cloud computing".

Source:
http://www.wired.com/2011/12/service-level-agreements-in-the-cloud-who-cares/

Week Five

Week Five

This week I decided to take a look into Internet Protocol version 6 (IPv6).

IPv6 is the next version of Internet Protocol (IP). It has been available since 1999, but is now slowly starting to replace IPv4. One website I found is a IPv6 fact sheet created by the Internet Society. I found this site interesting because it answers questions and gives the needed facts, all in a two page document. According to the fact sheet, IPv6 provides more than 340 trillion, trillion, trillion IP addresses, allows a huge range of devices to connect directly with one another, and helps ensure the Internet can continue its current growth rate indefinitely.

Even though 96% of the internet is still using IPv4 it is expected to run out of addresses within the next few years. IPv6 must be implemented to keep the internet as we know it operational.

An easy way to tell if you are ready for IPv6 is to use the Google test site:
http://ipv6test.google.com/

As an example this was the result from my work computer:

Source:
http://www.internetsociety.org/what-we-do/internet-technology-matters/ipv6
http://ipv6test.google.com/