p.p1 which services are providing and therefore doubling

p.p1 {margin: 0.0px 0.0px 0.

0px 0.0px; font: 12.0px Helvetica; color: #454545}p.p2 {margin: 0.0px 0.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now

0px 0.0px 0.0px; font: 12.

0px Helvetica; color: #454545; min-height: 14.0px}The internet started off as ARPANET, developed by the US DOD in 1969 by the Advanced Research Projects Agency (ARPA), operating in 4 locations at its conception. Today, ARPANET is the internet, spanning 200 countries across hundred of millions of locations.

Three forces have driven the evolution of networking: traffic growth, development of new services, and advances in technology.Traffic used to be limited to voice and data. Now, it includes images, video, and streaming. Business moving to the web, commerce, and social networking continue to push traffic’s limits. And as businesses move to the web, the services they provide are naturally more web-centric. Businesses also have to consider that their users interface with mobile phones as much as they do with personal computers (if not more), doubling the channels in which services are providing and therefore doubling the traffic. And of course, advances in technology such as more powerful computers lead to the support of more powerful applications; powerful applications not only demand a lot of processing power, but a lot of internet bandwidth to go with the processing power.

Because of this, ISPs double their internet backbone data rates every 18 months, and enterprise/LAN applications double in their data rate standard every 2 years. LANs, in the past, supported only light workloads for file transfer and electronic mail. Today, LAN users may need to access centralized server farms to download several gigabytes of data at a time. LANs are a part of practically every enterprise, as (large) file-sharing (presentations, demos, training videos), messaging, and accessing of data local to an office is more commonplace. Quickly, though, WANs became the norm in enterprises, as corporations with locations across continents became more interconnected.

Companies have begun to disperse into smaller and smaller offices as telecommuting becomes commonplace. Increasingly more traffic traverses the WAN now, and therefore, internet providers have to provide the backbone to support large file transfers that used to occur locally. And with the emergence of applications that support several different types of data (VOIP, email, messaging, video, CRM, ERP, etc.) at the same time, networks that used to have to deal exclusively with data traffic now need to support these multi-faceted applications. The number of internet users has more than doubled in the past 8 years, from 2 billion users in 2011 to 4 billion in 2018, supporting over 20 billion networked devices. With more devices per user, users are increasingly consuming network capacity more continuously (mobile phones are on even whilst users are asleep, consuming data as they receive and apply software updates, alerts, and messaging).

Total traffic (in bytes) has quadrupled in the same time-span (not only are devices increasing in numbers, but in the amount of data they consume). I used to work at a company called Talari Networks, which was an SD (software defined) WAN company. We sold devices to enterprises that aggregated their redundant internet connections so that they could use all the bandwidth they were getting from their connections at the same time. For example, if a company had 3 different ISPs, each providing a 100mbps connection, they now had 300mbps they could use simultaneously, whereas without a Talari device, they could only utilize one of these links at a time. Therefore, companies experiencing an issue with limited bandwidth/options in an area from single ISPs could now deploy a Talari appliance and solve that issue. This was a big problem for companies with sites in the middle of nowhere, where ISPs had limited bandwidth. Perhaps all you have in the middle of the desert is a satellite internet ISP, and a couple 3g ISPs. All relatively unreliable options.

With Talari, aggregating 3 ISPs @ 95% availability (1.5 days of downtime per month) results in ~98% availability (16 hours of downtime per month). A company might’ve had 3 different ISPs for redundancy — for example, 911 call centers (we had many customers in this space) would have several ISPs as if one went down or was experiencing a brownout, they would failover to the next one.

That said, many company had redundancy but their failover wasn’t as smooth as a 911 call center’s. Perhaps a network engineer might have to manually fail them over, or their automated failover took 1 or 2 minutes at times. With Talari, there is no failover. Links are always active-active, and therefore if one link goes down, packets are pushed into the next active link. So, call centers would never have to deal with dropped calls, even through outages.

Talari was solving issues of bandwidth limitations, as well as availability limitations. Our competitor, Riverbed, was doing WAN Optimization (caching and reduplicating data on each end of the network) to minimize the amount of data traversing a network. Combining WAN Optimization and SD-WAN (something companies like Citrix are currently doing) is the next generation of WAN technology. Demanding applications and behemoth-sized WANs will require the combination of these 2 technologies to deliver five-nines reliability in WANs. 

Author: Patrick Santiago


I'm Mia!

Don't know how to start your paper? Worry no more! Get professional writing assistance from me.

Check it out