Perimeter 4min de Leitura - 23 de November de 2015

Proxy cache: Content acceleration with caching techniques

Teclado de notebook

This post is also available in: Português English Español

The amount of content made available on the internet over time has become a major challenge for telecommunications companies and internet service providers. In the same way, in smaller scales, it has affected companies of diverse sizes and segments with respect to the consumption of band.

The number of Internet users grows very fast and, according to the Internet Society, we have surpassed 3 billion users. In 2012, approximately 50% of Internet traffic was video, and today, more than 50% of mobile phones are smartphones, which consume more and more content on the internet.

There is an immense amount of content, widely available to be accessed by the most varied devices. This has brought a great challenge to information security in terms of privacy and integrity, but also an incredible hurdle in the issue of availability.

As the speed of access to the Internet of companies is limited, depending on the number of users, it is common to have periods of greater access and consequent perception of delays (slowness) in the processing of requests and responses. This directly affects productivity in the company, especially in administrative areas where there is intense use of computers and internet.

This article introduces the concept of proxy cache, used to accelerate content through the internet, allowing companies to have a greater rationalization of the use of their links, and consequently productivity.

Content cache

There are several caching techniques, all of which have the common purpose of storing certain information for a period to be made available more quickly when requested.

Content caching can be considered a key element in a bandwidth savings and content acceleration architecture. Moreover, through the same structure, these two areas are fully met.

Bandwidth savingsare generated by caching since you do not have to go to the internet to download an object again. In addition, just for the same reason, there is a perception of acceleration, because instead of the download being carried out, the download is done locally.

There are several cache types that operate in a standalone and centralized architecture, providing faster access to requests made by users. We will be exploring low complexity features, closing with multipurpose accelerators.

Browser cache

As simple as it may seem, the caching feature in browsers acts as an acceleration mechanism, even though it is totally limited and cannot be shared with other users within a corporate network.

The feature present in browsers in conjunction with cache proxies acts complementarily in a caching architecture, allowing locally stored objects not even required by the proxy.

However, it is important not to have high expectations of acceleration by using browser cache, since many store the objects in memory, which are lost at the time the software is shut down. In cases of disk cache usage, the space allocated for this purpose is quite limited since such function it is not the purpose of the application.

Proxy cache

The proxy-integrated cache provides a centralized mechanism for storing content, which is shared with other users of a corporate network, so there is a much larger set of data stored, positively affecting content acceleration.

The more common accesses, the greater the caching and the lower the response time for the user within the company. Storage is a limited resource, however, with replacement policies the cache structure keeps the objects more accessed within the company’s context and access profile.

Many proxies offer caching function, but it can also be used as a proxy firewall, regulating traffic, allowing or blocking it according to the policies established in the solution itself.

Together with a browser cache, this structure offers fantastic acceleration gains and bandwidth savings, regardless of the type of content commonly used by the network and its users. Static content such as images, documents and the like, is easier to find while streaming and other dynamic content is more complex.

For dynamic content, specialized solutions exist for caching videos on Youtube, Netflix, Vimeo and others. These services offer particularities in the format of content distribution, which makes it difficult for a traditional cache to handle the storage of information in a structured way.

Given that heavy web traffic is streaming, a static Internet-only ISP will not have considerable savings or acceleration. The same is true for a company, so it is interesting to evaluate the need based on the traffic profile. A marketing and advertising agency, while not an internet provider, will certainly need a dynamic cache structure.

Multipurpose accelerators

On the other hand, leaving the web zone, there are multipurpose accelerators that act directly at the network level. These devices have medium to large business applications, helping to optimize WAN capabilities, especially when there are huge demands for data interconnection between different websites/datacenters.

Usually composed by dedicated hardware, this type of accelerator has a high cost of acquisition and for the correct operation must be added in all the tips that are part of the communication. This is because they have proprietary compression structures, and therefore need to be in all locations to deliver the benefit of bandwidth throttling and economy.

In addition, many accelerators also have in-depth knowledge of the behavior of certain applications, offering different types of effectiveness in terms of compression. For example, an engineering office that works with CAD projects, with geographically distributed units interconnected over the internet may have an accelerator in each unit that will offer compression rates of up to 60% of this type of application.

Another great point is that this type of accelerator does not normally operate as a proxy, which guarantees a high degree of transparency and continuity of environments in the face of changing protocol behaviors. In the case of proxies, if the supported protocol changes and is not properly updated, some new application or software may not work.

Content Delivery Networks

Still on the topic, the term Content Delivery Networks (CDN) is a great example of cache and acceleration structure that is increasingly present as a way to bring content closer to users.

The structure has a highly distributed composition, usually formed by large cache structures, stored in several datacenters with strategic geographic proximity to offer content to users with more speed and availability.

This model is very interesting for the content production company as well, as it can perform a work of offloading traffic in its structure, making the environment as a completely more scalable and resilient to denial of service attacks.

Start thinking about how to optimize the use of the internet in your company, using techniques and proxy cache, or some other method suited to your need. Did this article help you to understand a little bit about the topic of web content acceleration? Does your company already have technology for this purpose? Leave your comments and we will enrich this article.

This post is also available in: Português English Español