Browse Prior Art Database

High Capacity Scoped Leaderboards for Live Events

IP.com Disclosure Number: IPCOM000226945D
Original Publication Date: 2013-Apr-25
Included in the Prior Art Database: 2013-Apr-25
Document File: 4 page(s) / 169K

Publishing Venue

Microsoft

Related People

Darrin Brown: INVENTOR [+4]

Abstract

The objective was to provide custom leaderboards for a very large number of users in a very short timeframe. The leaderboards are user-specific since they only include a users friends from their social network (as opposed to a single global leaderboard with all users included). Leaderboard results need to be calculated and prepared for a user to view as soon as possible after a live event concludes. Only people who actually participated in the meta-game (e.g. picking game or event outcome winners) are included in the leaderboard rankings.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 60% of the total text.

Document Author (alias)

Darrin Brown (dabr)

Defensive Publication Title 

High Capacity Scoped Leaderboards for Live Events

Name(s) of All Contributors

Darrin Brown

Mike Perzel

Jackie Li

Usman Khalid

 

Summary of the Defensive Publication/Abstract

The objective was to provide custom leaderboards for a very large number of users in a very short timeframe. The leaderboards are user-specific since they only include a users friends from their social network (as opposed to a single global leaderboard with all users included).  Leaderboard results need to be calculated and prepared for a user to view as soon as possible after a live event concludes. Only people who actually participated in the meta-game (e.g. picking game or event outcome winners) are included in the leaderboard rankings.

The original solution indivdiuallly calculates the leaderboard on demand (i.e., per user request); this could not handle the traffic load and latency.

This solution is designed to handle 1 million users producing 1million distinct leaderboards within 2 minutes of receiving the event result that impacts the leaderboards outcome.  The current Xbox LIVE implementation limits each user to having a maximum of 100 friends (on their leaderboard).

 

Description:  Include architectural diagrams and system level data flow diagrams if: 1) they have already been prepared or 2) they are needed to enable another developer to implement your defensive publication. Target 1-2 pages, and not more than 5 pages.  

We separated our server architecture into Workers and Web Frontends. The job of the Web Frontends is to handle communicate between User Clients and additional Data Storage. The job of the Workers is to calculate leaderboards based on users data and event results. The communication between web frontends and the workers is using the Work Queue which is an Azure storage queue.

The Web Frontend has a number of functions that include storing and retrieving user choices from Data Storage.  We are interested in how the leaderboards are served up to the Users. The Web Frontent provides a manifest which contains links to the individual user’s leaderboard results.  These links point to a Leaderboard Storage which is just individual blobs stored in Azure blob storage. This is an example of the command query pattern.  This is beneficial because the leaderboards can be updated when and how th...