Blog

Stay up to date with the latest technological innovation at InMobi with the blog posts below.

Multi-GPU training of Large, Sparse-Matrix on Wide #NeuralNetwork

Author: Preetham V Vishwanatha, VP DataSciences ~ Deep LearningMulti-GPU training of neural network on TensorFlow (v0.12 as of this blog) is a pain. It is a pain if you get off the beaten path that is. I found out a bit early that the SparseTensors in the contrib.learn package does not play well on ...

Geo-location Hygiene

Location Accuracy: Who can you Trust?Thanks to the widespread availability of low cost GPS technology, many mobile devices have the ability to precisely geo-locate themselves. When users of these devices opt-in to make their location information available to applications and network services, a bett...

MT Raghunath and Ameya.Sakhalkar on September 25, 2016

Spinning up a New Colocation Facility!

Beyblade, the beginning!!!When you challenge yourself to light up a new data center built in 2 months, things look crazy. As per the comment of my program manager Masthan, “that sounds like spinning our DC-1 Data Center (DC) like a Beyblade, the toy my kids often play at home.” In action, this s...

Sanjay Kharb
Sanjay Kharb and Masthan on August 30, 2016

Building A Scalable Multitenant Metrics Service

As we discussed in part 2 of this blog series, a robust metrics service is critical to achieve convergence of metrics and monitoring. Considering the requirements discussed, we were clear that the service had to reside in our data centers for reasons of cost and control. Once we arrived at the archi...

Sudhakar BG and Bharath Ravi Kumar on May 16, 2016

A Converged System for Metrics and Monitoring

OverviewAs discussed in part 1 of the blog series, metrics are central to InMobi’s engineering methodology. Based on our requirements, it was clear that we needed a single source of reliable data for production metrics, to achieve convergence of monitoring, visualization, and analysis of productio...

Sudhakar BG and Bharath Ravi Kumar on April 27, 2016

Towards a Robust Infrastructure for Metrics Collection and Monitoring

[This is the first post in a blog series on our production infrastructure for metrics collection and monitoring. In this post, we discuss why metrics are critical to our engineering function, and the need for a robust infrastructure supporting metrics collection and monitoring.]Terminology & Taxonom...

Bharath Ravi Kumar and Sudhakar BG on February 29, 2016

UI Automation: Graph based Approach

A Graphical User Interface (GUI) is a fundamental component in the software we engineer today. It provides an interactive and easy connect to the underlying program making the overall software more usable, e2e workflows more passable and flexible. However, at the quality layer, GUIs pose a number of...

Abhish Agarwal
Abhish Agarwal on January 14, 2016

Testing Big Data Pipelines Made Easy With Apache Falcon

At InMobi, we see events arriving in excess of 10 billion per day. Analysis, reporting and inferencing from these requests (and responses served) is key to serving the right ad, to the right person, at the right time. We have nearly 200 complex big data pipelines that run against various data source...

Pavan Kolamuri
Pavan Kolamuri on November 12, 2015

Validating Hadoop Jobs at InMobi using Strider

Several technology companies run their analytics pipeline jobs on hadoop. Though easy to start, managing a scalable pipeline presents multiple challenges across different dimensions. In this article, we focus on one set of those important challenges -- the one arising from the functional testing per...

Swamynathan S
Swamynathan S on November 04, 2015

Real-Time Stream Processing at InMobi- Part 4

This is the fourth blog in the series of blogs regarding real time stream processing at InMobi. The first blog provided insight into the variety of company-wide use-cases that provided the motivation for stream processing at InMobi. The second blog provided a brief overview of both Storm and Spark...

Satish Mittal
Satish Mittal on October 14, 2015

Real-Time Stream Processing at InMobi- Part 3

This is the third blog in the series of blogs regarding real time stream processing at InMobi. The first blog provided insight into the variety of company-wide use-cases that provided the motivation for stream processing at InMobi. The second blog provided a brief overview of both Storm and Spark St...

Satish Mittal
Satish Mittal on October 05, 2015

Real-Time Stream Processing at InMobi- Part 2

This is the second in a series of blogs about real-time stream processing at InMobi.The first blog provided an introduction to the motivation behind stream processing at InMobi. It also gave insights into the number of company-wide use cases for stream processing.In this blog, we provide a brief ove...

Satish Mittal
Satish Mittal on September 28, 2015

Being an Efficient Mentee

In my last article I spoke about “Spotting the Right Mentor” as a core responsibility of a mentee. While this is very important to kick start the mentor-mentee relationship on the right note, there is a lot more that a mentee has to do to be efficient and make the engagement mutually enriching. ...

Mohit Saxena
Mohit Saxena on September 22, 2015

Real-Time Stream Processing at InMobi- Part 1

This is the first in a series of blogs that will detail the real-time stream processing at InMobi.IntroductionMany InMobi applications increasingly need to be able to process real-time data in a streaming fashion. The absence of a true streaming alternative has forced many of them to model themselve...

Satish Mittal
Satish Mittal on September 22, 2015

Understanding Cross Origin Resource Sharing (CORS)

Understanding Cross Origin Resource Sharing (CORS) is important from the security standpoint. The reason is explained below.Rich web experiences use a lot of Web API, CDN caching, and cross domain requests for resource sharing. Supporting CORS from the server not only helps in retaining maximum cont...

Ishan Girdhar
Ishan Girdhar on September 21, 2015

Creativity is a Bacon-Flavoured Beer

After having been really intrigued by the title of the song Happiness is a Warm Gun, I researched it a bit. The story goes that when he was shown the cover of a gun magazine that read ‘Happiness is a Warm Gun’, Lennon thought it was a fantastic, insane thing to say.“A warm gun means you just s...

Amit Deshpande
Amit Deshpande on September 03, 2015

Optimizing Joins in MR via Lookup Service

At InMobi, we process terabytes of data to produce understandable reports. These reports are queried for and analysed by our customers. Most of our raw data contains a lot of business keys that need to be joined with another set of data, generally referred to as dimension data, to produce enhanced...

Rohit Kochar
Rohit Kochar on September 01, 2015

Can You Spot The Right Mentor

“Coaches, mentors, gurus” – the roles people adorn themselves with, are often misunderstood in terms of the true meaning of what they entail, hence it continues to be a pet peeve of mine. I have been fortunate to be coached, rather groomed by a mentor who has made me who I am today. And I owe ...

Mohit Saxena
Mohit Saxena on August 25, 2015

Modeling Pipeline for User Segments

InMobi is an ad network, an entity that matches users to advertisers in a manner that maximizes efficiency. Doing this is a complex, multi-objective optimization problem. One has to cater to the preferences of several entities in the ecosystem - the publishers, the advertisers, and most importantly,...

Raghunandan M.A.
Raghunandan M.A. on July 20, 2015

ARPing a lot

While studying the traffic going to our core switches, we observed a high number of ARP packets getting into the Control Plane(CP) of the switches. The ARP broadcasts were very high and this was not an ideal behaviour under any circumstances. So, we wanted to know " What was the actual traffic hitti...

Brijesh KB
Brijesh KB on July 07, 2015

Handling DNS Lookups in a High Performance System

This article talks about building resilience and fault tolerance within the application against non reliable (or potentially blocking) external DNS lookups. It talks about moving the impact of slow external DNS lookups from the worker threads in the request path to asynchronous timer threads for enh...

Rajashekhar Choukimath
Rajashekhar Choukimath on June 23, 2015

Apache Lens Demystified

1. IntroductionThe Apache Lens Documentation describes Lens as a platform that provides an “Unified Analytics Interface” that provides a common view of data stored across multiple tiers and storage layers. In this article, we will try to break down that definition and try to explain, in layman t...

Jothi Padmanabhan
Jothi Padmanabhan on May 19, 2015

Managing a 14TB Reporting Data Warehouse

It was a great pleasure to be a part of first ever full day PostgreSQL event (PGDay India 2015 event) in Bangalore with more than 70 members attending it.For this event, I was given the chance of presenting on how we "Manage a 14TB Reporting Data Warehouse at Inmobi" using postgresql. To shed some...

Soumya Ranjan
Soumya Ranjan on April 22, 2015

24/7 Monitoring and Alerting of PostgreSQL

The digital universe is huge and is growing at a stellar rate and along with it grows the data generated every second. By 2020, there will be nearly as many digital bits as there are stars in this universe. That effectively means infinite as per the reports published by IDC in 2014. InMobi has grown...

Adarsh Sharma
Adarsh Sharma on April 20, 2015

Enterprise Grade Deployment and Security with PostgreSQL

PGConf US is organised by United States PostgreSQL Association every year to grow the PostgreSQL community. It's good gathering of PostgreSQL developers , different contributors , DBA , users and learners. PGConf US 2015 conference had 3 days session including new releases, features, live use cases ...

Himanchali
Himanchali on April 02, 2015

Beyond picking “two of three” from CAP

Citing the CAP theorem and the infamous “pick two of three” is a soundbite that we have all been subjected as a consequence of working on distributed systems. While the original Brewer’s conjecture gets a lot of mention, people often overlook his essay titled “CAP Twelve years later”. The ...

Arvind Jayaprakash
Arvind Jayaprakash on March 11, 2015

InMobi’s Big Data Analytics Platform Incubates into Apache

Analytics and Business Intelligence systems are important focus area at Inmobi. Hadoop based warehouses operate well over large datasets in the order of TBs and PBs and scales fairly linearly. Queries over Hadoop still typically adopts a full scan approach, which tends to be costly in terms of query...

Amareshwari Sriramadasu
Amareshwari Sriramadasu on February 23, 2015

What it’s like working in InMobi Technology and the PIE Team

WOW! Three letters that define my 4 months stint in the technology team at InMobi. It has been one of the most wonderful, enthralling and awesome time. There are endless opportunities and the passion to be the best in globe just flows down the veins in each and every team. Thought to pen down my vie...

Sanjay Kharb
Sanjay Kharb on February 15, 2015

InMobi at Hadoop Summit 2015

InMobi Technology Team submitted total 10 papers at Hadoop Summit this year. Here are the abstracts for them.1. Driving 10x Efficiency in Testing Hadoop Applications through AutomationAbstract: Production data pipelines keeps evolving rapidly as business and products evolves. One of the key challeng...

Saumya Verma
Saumya Verma on January 30, 2015

Apache Falcon just graduated !

Recently, Apache announced the graduation of Falcon to a Top Level Project. This is a milestone for the project and a testimony to the need of the hour for a robust big data processing and management solution for Apache Hadoop ecosystem. This project, which is now being widely used by various indust...

Pallavi Rao
Pallavi Rao on January 29, 2015