100%(1)1 out of 1 people found this document helpful
This preview shows page 9 - 11 out of 16 pages.
accomplish this goal, they need to be able to move large amounts of data over the network, handle multiple virtual machines (VM) and platform as a service (PaaS) product, and day to day internet traffic without causing slowdowns and congestions. These are the main areas we will focus on for identifying traffic estimates, analysis for peak traffic times, and identify any other issues that could produce networkslowdown. Once we have consolidated the analysis, we will identify recommendations of software and infrastructure that could handle the projected traffic and identify how Tres Comma AI can best manage network efficiency. In 2017 the average broadband user in the US consumed 190 gigabytes per month [ CITATION Fos17 \l 1033 ]. Tres Comma AI has 300 users at our largest office and using the above calculation Tres Comma AI could assume that on average they would consume 1.9 terabytes per day (190 Gb x 300 users / 30 days per month). This calculation could be on the low end as more of our employees will continue to bring other electronic devices to connect to our WAP and more data is transferred around our network and to the cloud. As Tres Comma AI moves more of their file repositories, automation, and hosting to the cloud we could see a drawdown in bandwidth being using inside our network to an estimated 300 gigabytes per day for our code handling needs. We will continue to see data traveling to and from the cloud as individuals pull and push to our code repository, but once the initial pull has been completed file transfers should be on the smaller side. In addition, when users pull new branches the cache from Git on there local machines should also help reduce data transfer needs. Most of the peak data transfer will happen at 0800 and 1600 as users pull down new code at the start of the day and push before close of business days, we encourage users to push and pull often but we estimate the bulk of thiswill still happen at open/close of business. Some of our smaller teams are not yet transferred to the
cloud and use our internal code repository as well as automation servers and we estimate that the number of teams and projects being worked on this shouldn’t cause any bandwidth issues. Tres Comma AI supports multiple operating systems for our clients and in order to best support these we have dedicated environments our developers can spin up to do any testing or development. Some of these environments are run on VMs hosted on our servers but some are run on PaaS products like Docker on the developer’s workstation. Docker allows developers to pull down operating systems, code bases and execute the code for automation, testing and performance debugging without having to worry about environment disparity across different machines and operating systems [ CITATION Doc20 \l 1033 ]. What this means is Docker could pull down a complete operating system, the entire code base and execute the code to include any calls to outside agencies. The most common Linux base operating system image coming in at 188 megabytes and Windows between 600 megabytes for a nanoserver all the way up to 9.3 gigabytes for windows server will all the support libraries [ CITATION Chr15 \l 1033 ].