cloud feeding data over skyline

Posted by & filed under List Posts.

Cloud computing is used by many organizations to bolster their systems during the holidays. The same benefits that cloud offers throughout the year become particularly compelling during the holidays, when traffic can surge and require resources that either are or are not delivered by your IT infrastructure. How can cloud give your company an advantage during the holidays and meet needs more effectively than can be achieved through traditional systems? What exactly are horizontal and vertical scaling? How specifically does cloud help improve performance? How can cloud testing help in preparation for peak periods?

Why cloud is so powerful for the holidays

Within a traditional model, there are two ways to go, essentially, as indicated by cloud management firm RightScale:

  • Underprovision – If you underprovision, you would be assuming the normal usage of the application at all times. You would be very efficient throughout all typical usage periods. The downside, though, would be that you would end up losing traffic when it would get busy because your capacity would be insufficient. You would be underprepared for peak periods such as the holidays. You could not keep up with the number of requests. Your credibility would suffer, as would your sales.
  • Overprovision – The other option is to launch resources to an extreme degree. You would be able to handle traffic all all times. You would be inefficient with resources, though, because during normal times you would have too many. You would be able to handle traffic throughout peak times such as the holidays, but your infrastructure would be needlessly costly year-round.

Cloud is a great option, and a better option, because of the way the technology is designed – to optimize scalability. It allows you to allocate and deallocate resources dynamically, avoiding the need to buy equipment in order to answer the higher number of holiday requests.

It also allows you to deliver high-availability. In a technological setting, availability is the extent to which a user is able to get access to resources in the correct format and from a certain location. Along with confidentiality, authentication, nonrepudiation, and integrity, availability is one of the five pillars of Information Assurance. If your data is not readily available, your information security will be negatively impacted.

In terms of scalability, since cloud allows you to scale your resources up and down as traffic fluctuates, you are only paying for the capacity you need at the time. You also have immediate access to sufficient resources at all times.

Cloud scalability – why this aspect is pivotal

Scalability is the ability of software or hardware to operate when user needs require its volume or size to change. Generally, scaling is to a higher volume or size. The rescaling may occur when a scalable object is migrated to a different setting; however, it typically is related to a change in the memory, size, or other parameters of the product.

Scalability means you can handle the load as it rises, whether you need a rise in your CPU, memory, network I/O, or disk I/O. An example would be the holidays, any time you run a promotion, or a situation in which you get unexpected publicity. Your servers may not be able to handle the sudden onrush of traffic. Your site or app will not go down, even if thousands of people are using your system at once, when you have enhanced your scalability with cloud. You keep selling and satisfying your customers.

Performance is a primary reason for optimizing your scalability. Scaling occurs in two directions, horizontal and vertical:

  • Horizontal – When you scale horizontally, that means you are adding more hardware to your ecosystem so that you have more machines to handle the load. As you add hardware, your system becomes more complicated all the time. Every server you add results in additional needs for backup, syncing, monitoring, updating, and all other server management tasks.
  • Vertical – With vertical scaling, you are simply giving a certain instance additional resources. This path is simpler because you do not need to do special setup with software, and the hardware is outside your walls. You can simply attach cloud servers that are already virtualized and ready to use.

Cloud’s impact on performance

Two key aspects of site performance are addressed by cloud, site speed and site uptime:

  • Site speed – Page load time, also known as the speed of your site, is one of the ways that your rank on search engines is determined. The speed of the site will impact how well you show up in search; but more importantly, it will improve your ability to meet the needs of those who come to your site. You will perform better based on small fractions of a second in the improvement of speed. There are many ways to speed up your site in conjunction with better infrastructure. Those tactics include getting rid of any files you do not need, using a strong caching strategy, removing extraneous meta data, and shrinking your images.
  • Site uptime — Your uptime is critical in getting you strong sales because you must have your site available to users in order for them to be able to look through products, figure out what they want to purchase, and place orders. When the site is not available, customers will get a 404 page in their browsers instead of what they want. That will mean you are not able to make the sale. It also means you suffer in search engine rankings, which are based in part on availability. You may not be able to sell as much in the future either, since it will frustrate a user to arrive at a 404 page and not be able to complete their shopping. You certainly do not want your site to ever go offline.

Cloud testing in advance 

The basic reason that ecommerce sites fail is that they are not tested appropriately. Companies do not always know how well their sites will perform when they have huge surges of traffic over the holidays in the absence of this testing. Sites would realize whatever performance issues they might have in their servers well ahead of time if they conducted the relevant testing.

To avoid these issues, you want to test. The traditional way that you would go about testing would be with hardware that you hardly ever need. The other option is to use cloud testing.

With cloud testing, you have independent infrastructure-as-a-service (IaaS) providers (aka cloud hosts) provide you simply manageable, scalable resources for this testing. You are able to save money on testing by using cloud hosting to simulate the traffic that an app or site would experience in the real world within a test setting. You can see how your site stands up when it is hit with certain loads of traffic, with various types of devices, according to the rules you establish.

There is another benefit of cloud testing too: it is closer to what your actual traffic model will be since it is outside your in-house network.

Adding cloud for better sales this holiday

Do you think bolstering your system with cloud servers might be right for you this holiday season? At Total Server Solutions, where we specialize in high-performance e-commerce, our cloud uses the fastest hardware, coupled with a far-reaching network. We do it right.