Website monitoring
Website monitoring is the process of testing and verifying that end-users can interact with a website or web application as expected. Website monitoring is often used by businesses to ensure website uptime, performance, and functionality is as expected.
Website monitoring companies provide organizations the ability to consistently monitor a website, or server function, and observe how it responds. The monitoring is often conducted from several locations around the world to a specific website, or server, in order to detect issues related to general Internet latency, network hop issues, and to pinpoint errors. Monitoring companies generally report on these tests in a variety of reports, charts and graphs. When an error is detected monitoring services send out alerts via email, SMS, phone, SNMP trap, pager that may include diagnostic information, such as a network trace route, code capture of a web page's HTML file, a screen shot of a webpage, and even a video of a website failing. These diagnostics allow network administrators and webmasters to correct issues faster.
Monitoring gathers extensive data on website performance, such as load times, server response times, page element performance that is often analyzed and used to further optimize website performance.
Purpose
Monitoring is essential to ensure that a website is available to users, downtime is minimized, and performance can be optimized. Users that rely on a website, or an application, for work or pleasure will get frustrated or even stop using the application if it is not reliably available.[1] Monitoring can cover many things that an application needs to function, like network connectivity, Domain Name System records, database connectivity, bandwidth, and computer resources like free RAM, CPU load, disk space, events, etc. Commonly measured metrics are response time and availability (or uptime), but consistency and reliability metrics are gaining popularity. Measuring a website's availability and reliability under various amounts of traffic is often referred to as load testing.
Website monitoring also helps benchmark the website against the performance of a competitors to help determine how well a site is performing. Website speed is also used as a metric for search engine rankings.[2]
Website monitoring can be used to hold web hosting providers accountable to their service-level agreement. Most web hosts offer 99.9% uptime guarantee and, when uptime is less than that, individuals can be refunded for the excessive downtime. Note that not all hosts will refund individuals for excessive downtime so one must become familiar with the terms of service of their host.[3]
Most paid website monitoring services will also offer security features such as virus and malware scanning which is of growing importance as websites become more complicated and integral to business.
Internal vs. external
Website monitoring can be done from both inside and outside of a corporate firewall. Traditional network management solutions focus on inside the firewall monitoring, whereas external performance monitoring will test and monitor performance issues across the Internet backbone and in some cases all the way to the end-user. Third-party website performance monitoring solutions can monitor internal (behind the firewall), external (customer-facing), or cloud-based Web applications.
Inside firewall monitoring is done by special hardware appliances which can help you determine if your internal applications’ sluggish performance is caused by: design of applications, internal infrastructure, internal applications or connections to any public Internet.
External performance monitoring is also known as end-user monitoring or end-to-end performance monitoring.
Real user monitoring measures the performance and availability experienced by actual users, diagnoses individual incidents, and tracks the impact of a change.
Measures of website availability
| Availability | Nines | Downtime per year |
|---|---|---|
| 90% | 1 nine | 876 hours |
| 95% | – | 438 hours |
| 99% | 2 nines | 87 hours, 36 minutes |
| 99.9% | 3 nines | 8 hours, 45 minutes, 36 seconds |
| 99.99% | 4 nines | 52 minutes, 33.6 seconds |
| 99.999% | 5 nines | 5 minutes, 15.36 seconds |
| 99.9999% | 6 nines | 31.68 seconds |
Various types of protocol
A website monitoring service can check other internet protocols besides HTTP pages and HTTPS such as FTP, SMTP, POP3, ActiveSync, IMAP, DNS, SSH, Telnet, SSL, TCP, PING, UDP, SOAP, Domain Name Expiry, SSL Certificate Expiry and a range of ports. Monitoring frequency occurs at intervals of once every 4-hours to every 15-seconds. Typically, most website monitoring services test a server, or application, between once-per hour to once-per-minute.
Advanced monitoring services capture browser interactions with websites using macro recorders, or browser add-ons such as Selenium or iMacros. These services test a website by running a web browser through a typical website transaction (such as a shopping cart) in order to check for user experience issues, performance problems, and availability errors. Browser-driven monitoring services detect not only network and server issues, but also webpage object issues (such as slow loading Javascript, or third-party hosted page elements).
An implementation of time performance monitoring for the Apache HTTP Server is the mod_arm4 [4] module.
Various types of monitoring
Users of website monitoring (typically network administrators, web masters, web operations personnel) may monitor a single page of a website, but can also monitor a complete business process (often referred to as multi-step transactions).
Servers monitoring from around the globe
Website monitoring services usually have a number of servers around the globe – South America, Africa, North America, Europe, Africa, Asia, Australia and other locations. By having multiple servers in different geographic locations, monitoring service can determine if a Web server is available across continents over the Internet. Some vendors claim that the more locations the better picture on your website availability while others say that three globally distributed stations are sufficient and more stations do not give more information.
Types
There are two main types of website monitoring
- Synthetic monitoring also known as active monitoring, and
- Passive monitoring also known as real monitoring.
Notification options: alerts
As the information brought by website monitoring services is in most cases urgent and may be of crucial importance, various notification methods, often known as "alerts" are used: e-mail, IM, regular and cell phones, SMS, fax, pagers, Skype, RSS Feed etc.
See also
References
- Amazon Downtime Cost $66,000 - Forbes
- Speed Affects Website Usage - Google Research Blog
- ↑ Stevenson, Seth. "What You Hate Most About Waiting in Line". Slate. Retrieved 25 October 2014.
- ↑ Costill, Albert. "SEO 101: How Important is Site Speed in 2014?". Search Engine Journal. Retrieved 25 October 2014.
- ↑ Tryn, Joe. "For newbies : 99.5% / 99.9% uptime guarantee explained.". Retrieved 29 October 2014.
Also, do know how committed a host is in honouring their guarantee. Do they just mention it in their marketing channels, eg websites etc. Or do they really specify it clearly in their “terms of service” (ToS) ? What kind of compensation they plan to honour if they exceed the agreed upon max downtime ? Not all host are willing to compensate for exceeding the max downtime. Honest and responsible hosts will refund the hosting fees you paid for a particular month upon you reporting to them that they had exceeded the max downtime - and this is stated clearly in their ToS.
- ↑ Apache ARM 4 module