Holiday Spikes on Retail Data Centers

Holiday Spikes on Retail Data Centers 10000 2011 2012 2013 Holiday sales increased Holiday sales increased Holiday sales are expected to increase 5.1% to billion 3.5% to billion 3.9% to A billion $4360.2 $0795 $002,1 Of these sales, e-commerce will account for $83 billion of total holiday sales -15% OF THESE SALES, MOBILE COMMERCE WILL REACH 4168 IN 2013 holion How do holiday sales impact data centers? Total Sales Online Sales raw transactions going through registers on the base network management of e-commerce transactions credit card processing credit card management for transactions i Data Centers account information communications networks marketing eblasts and inventory management customer communication FOR SOME RETAILERS, 20%-40% THE HOLIDAY SEASON REPRESENTS OF ANNUAL SALES IN 2012, HOLIDAY SALES REPRESENTED 19% OF TOTAL RETAIL INDUSTRY SALES What effect does this holiday demand have on infrastructure within retail data centers? traffic spike Example holiday online traffic loads traffic spike O Predicted demand 9 Actual demand Infrastructure Time November Ist Cyber Monday Two Weeks before Christmas Increases in e-commerce traffic can cause spikes that are above the capacity of the system leading to down time and unhappy customers. This can also have an impact on sales numbers if a retail site cannot preform to traffic demands. Down Time Unhappy Customers $ Sales Major retailers were unable to predict sales trends for the holiday season and were ill-prepared to handle the increased traffic, leading to lost sales. EXAMPLES OF IMPACT Holiday Season Holiday Season 2012 2011 Large technology Large undergarment Large clothing retailer retailer had an retailer had 80 was overloaded with traffic causing the site to go down for several hours during hour-long checkout minutes of down time process because of an during black friday application error product launch Tips to ensure your system can handle spikes from holiday traffic Measure and retain historical Yesterday NOW Future network data – daily network traffic load and traffic peaks Predict network utilization and build for growth of peak utilization Establish a 100% Visibility Network Monitoring Strategy Build in 2Q/3Q to be ready for 4Q traffic loads Build the Network Monitoring network and tools to manage the peak load and provide visibility window Utilize packet deduplication, packet slicing and load balancing to limit the network monitoring load and maximize visibility 2012 2013 next year Measure and analyze for next year's peak season SOURCES http://www.nrf.com/modules.php?name=Pages&sp_id=1140 http://artides.chicagotribune.com/2013-11-11/business/sns-rt-us-usa-retail-holidays-analysis-20131110_1_u-s-retailers-holiday-season-holiday-sales http://www.wirelessweek.com/artides/2013/11/holiday-season-2013-capitalizing-mobile-and-cloud http://blog.onr.com/data-center-2/use-data-center-to-prevent-christmas-shopping-website-outages/ APCON http://www.greenhousedata.com/blog/post/how-to-handle-seasonal-web-traffic-spikes-with-the-doud/ Solutions for Networks http://www.rightscale.com/info_center/white-papers/2013-cloud-for-the-holidays.pdf http://www.internetretailer.com/2013/09/05/online-holiday-forecast-sales-will-increase-151 apcon.com http://www.xo.com/newsletter/ 1309/?utm_campaign=Newsletter_Prospects_Sept2013&utm_medium=email&utm_source=Eloqua&ListSource=TheConne ction_Sept2013&elq=5a79ebe487bb42dca8b872efa08c3974&elqCampaignld=152 http://www.datacenterknowledge.com/archives/2013/11/29thanksgiving-day-traffic-surprises-performance-slows/your-home/pictures/indexhtml Demand