LSPE India

We completed our 41th event yesterday, 22nd Feb 2022. It was another virtual meet with speakers from AWS, Nvidia and Samsung Research sharing their expertise with the community. Here are some pics (screenshots now a days) :

Friends we had our #37th event, 1st event of our 10th year journey, today 22nd May 2021. Its pandemic time and we were virtual. The event was hosted by Atlasian, Bangalore. Here are few captures from the zoom meeting room :

Friends,

Fighting the nCOV-2 virus needs lots of computing. Folding@home (https://foldingathome.org/) is a novel distributed computing project running out of Washington University. The aim of the project is to create programs which Internet users can install on their personal computers and share their computing cycles and contribute in solving protein folding problems associated with COVID19 and other similar virus research, specially towards drug design. More details on how to participate is available here : https://foldingathome.org/start-folding/ .

The community is meeting today for its 32nd event (first virtual) over a Zoom video conference sponsored by Grab. Those who did not RSVP and would like to watch, please follow live stream here : https://www.youtube.com/watch?v=2Y8AaCsOTY8

When: Saturday, September 12, 2015 10:30 AM Where: Yahoo Software Development India Pvt Ltd.,“Torrey Pines”, Embassy Golf Links Business Park, Off Indiranagar- Koramangala  Intermediate Ring Road, Domlur, Bangalore

Event Agenda and Speaker info

10:45 - Registration

11:00 - Tea

11:15 - Opening and stage setting for 15th event.

11:30 - Simoorg: Failure Simulation using python in a cluster for testing the   stability of the code - Arjun Shenoy is a Site Reliability Engineer at LinkedIn India   Abstract : Simoorg the new failure simulation tool developed at LinkedIn.   It is a failure induction tool (similar to chaos monkey) . The main rationale behind developing the tool was to have an extensible and easy to   code framework for inducing failures. Failure Inducer works by introducing a particular set of failures to a healthy cluster and then reverting them after a period of time during which it the state of the cluster is logged. The talk is going to cover introduction to Simoorg, its key features that separates it from chaos monkey architecture and a demo. git url :​ https://github.com/linkedin/simoorg   About Arjun : https://in.linkedin.com/in/arjunshenoylinkedin  

12:15 - What is Your Monitoring Strategy - Aveek Misra , Devops Architect, Intuit  

In this talk Aveek would discuss about how the monitoring strategy has evolved given the changes in web applications and infrastructure. With the advent of public and private clouds and containers, the old notions of infrastructure monitoring are no longer that useful. The application landscape has also changed with monolith applications being split up into micro services and written in languages such as Node.js. Also many of the companies today are “Mobile First” which means that the monitoring of apps in mobile devices has now become very important. My talk will focus on some of these changes that are taking place in the monitoring landscape today and how your monitoring strategy should factor this in while setting up your monitoring.

Aveek is a DevOps architect in Intuit with more than 16 years of experience. For the past 6 years, I have been working in the Monitoring domain. In my past experience working with Yahoo, I have been very fortunate being part of the development team that created a highly scalable monitoring platform.

1:00 - Lunch

2:00 - Java 8 - New APIs and simplified parallelism- Deepak Shiwani, SDE3, Flipkart

Introduced in 1995, Java is programming language of choice for 9 million developers and today powers 7 billion devices. In March 2014, Oracle announced availability of Java SE 8, and we will see Java SE 9, released in 2016. Despite being in market for around 20 months, developers have not started extracting the best out of new features in Java 8 like lambdas, default methods, functional interfaces, streams apis etc.   In this talk, Deepak will start discussing about these new features specifically - the new streams API and ability to make the code parallel to utilize all the cores on the hardware more efficiently. Parallel programming was always there in Java prior to Java 8, but it involved lot of boiler plate code around Threads, Executors etc, but with Java 8, all this has been made easy. A more appealing reason to make your code faster. The session will include demo for sample code, written in traditional manner and then making it parallel using Java 8 constructs and that too - with minimal coding.

Deepak Shevani is SDE 3 at Flipkart, and works with teams building next generation of e-commerce platforms. Prior to Flipkart, he has worked at Yahoo as part of search team, where his focus areas were image and video search algorithms. He is keen to meet folks across industry and share insights about the trends and upcoming technologies.

2:35 - Public vs Private Cloud, a comparison of Openstack with AWS with a short demo - Ajeesh Rao and Ashutosh Sharma, both Chief Engineer, Samsung Research Institute, Bangalore  

(Abstract would follow)

3:10 - Microservices with Docker - a quick howto - Ravi , Cisco

3:45 - Tea, snacks and networking.  

Parking directions

LSPE-IN in collaboration with Devops Bangalore Meetup community is proud to present the next edition of our quarterly meeting    When: Saturday, September 12, 2015 10:30 AM    Where: Yahoo Software Development India Pvt Ltd.,“Torrey Pines”, Embassy Golf Links Business Park, Off Indiranagar- Koramangala  Intermediate Ring Road, Domlur, Bangalore

Event Agenda and Speaker info

10:45 - Registration 11:00 - Tea 11:15 - Opening Note : Measuring Computing - how lspe-in can partner with CMG and build a stronger community - Milind Hanchinmani 11:30 - Experience of designing a NoSQL backend for Keystone (Authorization and authentication component of Openstack) - Ajaya Agrawal, Devops, Reliance

Abstract: OpenStack Identity, or Keystone, is the project that provides authentication and authorization for use by all projects in the OpenStack family. It provides the APIs and the backing implementation to manage key areas such as identity, resources, and assignment. The default backend for persisting data for this purpose is MySQL, deployed either in a master-slave mode, or multi-master mode. In master-slave mode, the master is a single-point-of-failure. Moreover, the slaves are eventually consistent with the master due to the replication lag. A multi-master deployment such as that offered by Galera, improves availability as well as request handling capacity. The price incurred for this improvement is the increased latencies for every write. Moreover, this solution only scales up to the capacity of a single node, but does not scale out. To help with large deployments, we added an optional NoSQL backend based on Cassandra to address both concerns of single-point-of-failure, and lack of ability to scale-out.

12:15 - Azkaban-Scheduling Hadoop and BigData jobs - Arpit Tak, Software Engineer ,  Vizury

Abstract : This session will be covering Azkaban overview with its unique features using in large data systems. I will get more details into its EcoSystem, Architecture, Scheduling Multiple Hadoop Jobs, Detecting and Resolving Hadoop Job Failures, Resolving Hadoop Job Dependencies and internals of Azkaban.

Arpit works in Vizury as Software Engineer in Analytics Team and have around 2 yrs of experience in this domain

1:00 - Lunch

2:00 - Platform as a Service @ Yahoo - Sakthivel S, Senior System Engineer , Yahoo

Abstract:We at yahoo media thrive for excellence and deliver amazing customers experience . To provide excellent service, we developed Platform as a Service (PaaS) hosting infrastructure which allows us to provision, deploy, run and scale applications tremendously easy, fast and reliable.  The team will be sharing their experience of transiting from bare-metal  to PaaS via IaaS. Also we get a bird eye view of implementing PaaS.

Sakthivel is working as a Senior Systems Engineer at Yahoo handling critical properties like Frontpage and News

3:00 - Taming Graphite for a million metrics a minute - Jaseer T. K , Senior Operations Engg, InMobi

Abstract : Jaseer will walk you through how we designed and implemented monitoring as a shared service, at InMobi. This talk covers why we chose Graphite, how we benchmarked it and how we designed the infrastructure.

Jaseer is a Senior Operations Engg at InMobi, works on automation and managing monitoring infrastructure

 3:45 - Open forum

4:00 onwards - Tea, snacks and networking.

Parking directions

LSPE-IN in collaboration with Devops Bangalore Meetup community is proud to present the next edition of our quarterly meeting    When: Saturday, June 13, 2015 10:30 AM    Where: Yahoo Software Development India Pvt Ltd.,"Torrey Pines", Embassy Golf Links Business Park, Off Indiranagar- Koramangala  Intermediate Ring Road, Domlur, Bangalore

Event Agenda and Speaker info

10:45 - Registration 11:00 - Tea 11:15 - Opening and keynote

Internet of Things: Implications on Production Engineering - Radhakrishna Ayyappanaicker, Director- Production Engineering Yahoo! As a PE and OPs personnel, our goal has been upkeep and running of production systems as smooth as possible. While this goal has been same for years, no change in it, but we had to continuously evolve the way we do things and adapt to new technologies as system needs to scale further. This would become much more complex with smart phones and mobile technology increasingly becoming available to a larger number of people and a sudden spurt in growth of internet of things in our daily lives. Let us get the glimpse of what that trend is all about and what are the challenges ahead for us in building and managing reliable systems at scale.Radha has been working in IT industries close to 17 years, worked on both software development and production engineering roles. He has been with Yahoo for about 11 years. He is part of Advertising and Data platforms group, leading production engineering to upkeep and running of large footprint of Oracle/MySQL database clusters and Data analytic systems.  11:30 - Managing Openstack with Puppet - Sanay, Devops, Reliance

We use Openstack to manage bunch and bunch of VMs with bunch of hypervisors. How about managing Openstack itself with a popular CM system ? Sanjay does the same at his workplace. In his talk he would explain how he does it with a demo to fire your mind and set you up in experimenting path.

Sajay : Sanay works at Reliance Jio as a devops guy. He has been busy in laying out the infra for the Reliance Jio Cloud setup and mostly engaged with openstack provisioning via puppet. With +15 years of Sys Admin background he has worked with large setups including managing multimedia search frontends and Hadoop clusters at Yahoo.  I am father to three kids and love traveling by road.

12:15 - Apache Helix: Simplifying Distributed Systems - Shahnawaz Saifi, SRE @ LinkedIn Apache Helix is a generic cluster management framework used for the automatic management of partitioned, replicated and distributed resources hosted on a cluster of nodes. Helix automates reassignment of resources in the face of node failure and recovery, cluster expansion, and reconfiguration. In this talk Shahnawaz will cover Helix Introduction, concepts and putting concepts together to work. Shahnawaz is part of Site Reliability Engineering - Distributed Data Systems at LinkedIn. He has 6+ years of experience playing around large scale environments. Prior to LinkedIn, he was associated with Clickable and Guavus. He believes in "When you get, Give. When you learn, Teach!" 1:00 - Lunch

2:00 - Processing large data in near real time with Apache Spark - Ravi, Tech Lead, Cisco In this session, Ravi will cover Apache Spark overview with its unique features using in large data systems. I will get more details into Spark EcoSystem,Architecture, Elements and comparison with MapReduce. He will also touch up on its languages support with working demo session.

Ravi is in in IT industry for 11+ years. Ravi works for Cisco as Technical Leader and part of Cisco service team .He completed MS from BITS and BE from University of Madras. He has well  experience in building highly distributable systems using multi-tier architecture. His interest on exploring new technologies and tools.

3:00 - Voldemort - A distributed key-value storage system - Akhil Ahuja, SRE @ LinkedIn In this session Akhil will cover Voldemort overview and architecture. All aspects regarding replication, consistency, availability will be covered. The talk will end with short demo of important aspects of setting up a sample instance of voldemort.   Akhil works with LinkedIn as a Site Reliability Engineer for the past year. He has joined the distributed data storage team @ LinkedIn immediately after passing out of college. He has previous internship experiences with Juniper networks and Amazon. He has completed B.Tech in Information and Communication Technology. His interests are to know in-depth the various data solutions available out there and what distinguishes them from each other.  3:45 - Open forum

4:00 onwards - Tea, snacks and networking.

Parking directions

Enter into the EGL park from Koramangala Indiranagar intermediate ring road. The second building on your left is the venue, Yahoo!. This is the Torrey Pines building.  Ask the security for help and tell them that you have come for the #LSPE-IN event at Yahoo and they will help you park and entry.

For the first time in the history of #lspe-in and Devops Bangalore Meetup communities, we have decided to come up with a collaborative event. 13th June 2015, the 13th #lspe-in event would become a joint Devops Bangalore and #lspe-in event.   

We, the organizers of these two communities, felt that there are common areas of interest between us. While the #lspe-in focuses on technology challenges around scale the devops group targets to address the cultural shift needed in our thought process to scale well .   

The joint session would take place at Yahoo, Torrey Pines building . The event would be announced in both the groups however the registration may take place from one place . We would come up with more details in both the groups.

Friends,

Our next event would take place on 14th March 2015. If you have a talk proposal, please let us know at https://meetup.com/lspe-in. You may also mail us at [email protected] .

The Free Software Foundation, core behind the GNU Project, turn 30 in 2015. The FSF provides critical infrastructure and funding for the GNU project, the foundation of the popular GNU/Linux family of free operating systems and the keystone of the Internet.