Loading…
This event has ended. View the official site or create your own event → Check it out
This event has ended. Create your own
Get more information on MesosCon Asia, or register to attend.

Customize your schedule by experience level and/or presentation language:

Refer to the “Filter by Type” list on the right to find a session based on topic and/or experience level.

Presentation Language – Sessions are categorized as [C] Chinese, [C,E] Chinese with English Slides or [E] English at the end of each talk title.

View analytic

Sign up or log in to bookmark your favorites and sync them to your phone or calendar.

Tuesday, June 20
 

10:00

Hackathon
The focus, while not excluding other ideas, of this hackathon is improving Apache Mesos and it ecosystem. There will be a number of experienced contributors and committers helping the participants to get started, give direct feedback and if possible even commit code.
Potential areas include:
  • Documentation and website 
  • Apache Mesos Core (we will provide a number of tagged beginner issues)
  • Frameworks and DC/OS SDK
  • Integration (e.g., Cloud Provider templated)
  • Apache Mesos UI 
  • CLI
  • Benchmarking
  • Monitoring 
  • Tooling (e.g., local vagrant or docker setups)
All other proposals are welcome as well and will be discussed at the beginning of the hackathon.

Tuesday June 20, 2017 10:00 - 16:00
Room 305
 
Wednesday, June 21
 

08:00

Registration
Wednesday June 21, 2017 08:00 - 17:15
CNCC

09:00

Keynote: Apache Mesos Review and Vision - Ben Hindman, Founder, Mesosphere
Speakers
avatar for Ben Hindman

Ben Hindman

Founder, Mesosphere
Benjamin Hindman is a Founder and Chief Architect at Mesosphere where he leads a team building out core services for the Mesosphere Data Center Operating System (DCOS). Ben co-created Apache Mesos as a PhD student at UC Berkeley before bringing it to Twitter where it now runs on... Read More →


Wednesday June 21, 2017 09:00 - 09:25
Room 309A

09:25

Keynote: Update in China Unicom adoption with DC/OS [C] - XiangDong Geng, VP & GM, China Unicom
In this keynote talk, China Unicom will update the latest status of best practices in using DC/OS with their production. They will share what kind of applications in DC/OS, and how to leverage the advantages of DC/OS in their production with thousands of nodes.

Speakers
XG

XiangDong Geng

VP & GM, China Unicom
VP & GM of China Unicom


Wednesday June 21, 2017 09:25 - 09:45
Room 309A

09:45

10:05

Keynote: Apache Mesos and Standardizing Container - Jie Yu, Software Engineer, Mesosphere
Containers are now everywhere. Apache Mesos, as one of the most powerful container orchestrators, greatly simplifies the deploy, provision and execution of containerized workloads.  In this talk, I will talk about the evolution of container technology in Mesos. In particular, how Mesos embraces the industry standard for container networking, storage and image specification, and how Mesos achieves that by using a pluggable and extensible architecture.

The first part of this talk will give you an overview about the container technology in Mesos and how it has evolved over the years. Then, I will dive into three specific areas in the container technology: networking, storage and image provisioning, and the three industry standards that Mesos is adopting: CNI (Container Network Interface), CSI (Container Storage Interface) and OCI (Open Container Initiative) image spec.

Speakers
JY

Jie Yu

Tech Lead, Mesosphere
Jie Yu is a Tech Lead at Mesosphere, Inc, focused on Mesos development. He is a long time Apache Mesos committer and PMC member (since 2014). He leads the development of key container, networking and storage projects at Mesosphere. Before joining Mesosphere, he was a software eng... Read More →


Wednesday June 21, 2017 10:05 - 10:30
Room 309A

10:30

Keynote: Low-Latency In-Memory Computing and Its Implication on Scheduling [C,E] - Haibo Chen, Professor; Director of Institute of Parallel and Distributed Systems; Vice Director of Shanghai Key Laboratory of Scalable Computing and Systems, Shanghai Jiao
In-memory computing promises 1000X faster data access speed, which brings opportunities to boost many traditional computing into a higher level. In this talk, I will describe our recent research efforts in providing speedy in-memory transactions/queries at the scale of millions of transactions/queries per second and sub-millisecond latency. Moreover, I will also discuss the implication of low-latency in-memory transactions/queries on large-scale scheduling and outlines future work.

Speakers
avatar for Haibo Chen

Haibo Chen

Institute of Parallel and Distributed Systems, Shanghai Jiao Tong University
I have been a Full Professor in School of Software, Shanghai Jiao Tong University since December 2011. I currently lead the Institute of Parallel and Distributed Systems and work with members to improve the performance and dependability of computer systems. Before that, I was a f... Read More →


Wednesday June 21, 2017 10:30 - 11:00
Room 309A

11:00

11:30

Libprocess: The C++ Concurrent Systems Library Underpinning Mesos - Jay Guo, IBM China & Benjamin Mahler, Mesosphere
Libprocess is an asynchronous C++ programming library to support building systems out of composable concurrent components. Mesos is implemented using Libprocess and as such knowledge of Libprocess is essential when making significant contributions to the Mesos source code. This talk will provide a high level overview of Libprocess, including the motivation for the library, the core building blocks it provides, how it compares to other approaches, some examples, and more.

Speakers
JG

Jay Guo

IBM, Software Developer
Jay Guo is a Software Engineer at IBM China. He is an active Apache Mesos contributor and has been given several talks on MesosCon, Mesos meetups and container tech events.
avatar for Benjamin Mahler

Benjamin Mahler

Senior Software Engineer, Mesosphere
Benjamin Mahler is a committer and PMC member of Apache Mesos and has been working on Mesos since 2012. Benjamin now works at Mesosphere as a technical lead and has given Mesos related talks at several conferences and companies. His interests include distributed systems, fault to... Read More →



Wednesday June 21, 2017 11:30 - 12:15
Room 309A

11:30

Nightmares of an Apache Mesos Support Engineer - Jörg Schad & Jan Repnak, Mesosphere
A lot of talks will tell you how to setup a systems correctly. This talk is about what not to do with your Apache Mesos and DC/OS cluster.

We will share some of our favorite/scariest support stories covering typical system-setup, configuration, and application pitfalls for new (and not-so-new) Mesos and DC/OS operators. And, we will give some hints about how to debug those pitfalls if you do encounter them, resulting in fewer nightmares.

Speakers
JS

Jörg Schad

Software Engineer, Mesosphere
Jörg is a software engineer at Mesosphere in Hamburg. In his previous life he implemented distributed and in memory databases and conducted research in the Hadoop and Cloud area. His speaking experience includes various Meetups, international conferences, and lecture halls.



Wednesday June 21, 2017 11:30 - 12:15
Room 309B

11:30

Mesos at CMCC - Yuanzheng Chen, China Mobile
In IaaS age,the base platform in CMCC has many problems,such as inefficient deployment,low resource utilization rate,which leads to high labor cost in Application intergration and high devices cost.In this presentation,Chen Yuanzheng will introduce the PaaS platform in CMCC,key application refactoring process and look forword to future works.

Speakers
avatar for Yuanzheng Chen

Yuanzheng Chen

DCOS Architect, CMCC
Chen Yuanzheng graduated from Shanghai Jiao Tong University with a master degree in EE and is a DCOS architect at CMCC,the world's largest mobile phone operator. His study interest focuses on container management and resource utilization improvement.He is also responsible for app... Read More →



Wednesday June 21, 2017 11:30 - 12:15
Room 308

12:20

Debugging Tasks with Attach/Exec in Apache Mesos - Jie Yu, Mesosphere
Have you ever wanted to run the equivalent of `docker exec` or `docker attach` for tasks running in Mesos? Now you can. In this talk I will introduce how we added support for these commands to Mesos and how you can leverage them to more effectively debug containers running in production.

Speakers
JY

Jie Yu

Tech Lead, Mesosphere
Jie Yu is a Tech Lead at Mesosphere, Inc, focused on Mesos development. He is a long time Apache Mesos committer and PMC member (since 2014). He leads the development of key container, networking and storage projects at Mesosphere. Before joining Mesosphere, he was a software eng... Read More →


Wednesday June 21, 2017 12:20 - 13:05
Room 309A

12:20

How We Handle Tens of Billions of Rich Media Processing Requests Per Day - Wei Wei, Qiniu
Mesos is at the core of QINIU's rich media data processing system.  In this presentation, the presenter will talk about how QINIU build a PAAS platform based on mesos to support the large scale rich media data processing system.

First, the service scenario and challenge they faced, Second, the scheduler used to deploy and maintenance the rich data processing service, the service discovery, monitoring and logging subsystems. I will also talk about several lessons learned through this jounery and future plans.

Speakers


Wednesday June 21, 2017 12:20 - 13:05
Room 308

13:05

Lunch
Wednesday June 21, 2017 13:05 - 14:20
Level 3 Foyer

14:20

Secrets Management in Mesos [E] - Vinod Kone & Gregory Mann, Mesosphere
Without first class support for secrets (Credentials, API keys etc), application and framework developers had to traditionally resort to out-of-band mechanisms to transmit and use secrets in Mesos. This is changing now!

In this talk, we will describe the newly-added native support for secrets in Mesos. The Secrets API will let Mesos inject sensitive information into either the container environment or sandbox without exposing it to unauthorized actors. The advantage of this approach is that users do not need to explicitly store sensitive information in their task configuration but can let Mesos retrieve it from a secret store at container launch time. This is possible via the new “Secrets Fetcher” module interface which allows operators to integrate Mesos with 3rd party secret stores (e.g., HashiCorp’s Vault).

We will show a cool demo at the end that shows secrets in action.

Speakers
avatar for Vinod Kone

Vinod Kone

Distributed Systems Engineer, Mesosphere
Vinod Kone is a committer and PMC member of the Apache Mesos project. He is currently a Tech Lead and Engineering Manager @ Mesosphere. Previously, he was a Tech Lead and Manager of the Mesos team @Twitter. Vinod completed his PhD in Computer Science from UC Santa Barbara.
GM

Gregory Mann

Greg Mann is a distributed systems engineer at Mesosphere and an Apache Mesos contributor. He came to distributed systems by way of high performance computing, performing computational studies of solid state material properties during his time in the physical chemistry PhD progra... Read More →



Wednesday June 21, 2017 14:20 - 15:05
Room 309A

14:20

Jupyter and Spark on Mesos: Best Practices [C,E] - Shuai Lin, Scrapinghub
Many companies make use of Apache Spark and Apache Mesos to do offline data analytics. Before launching these jobs to production, it is important to provide a place for engineers to test their spark code interactively. At Scrapinghub, we have set up a Jupyter notebook server for this purpose. Engineers can launch their spark jobs via Jupyter interactively on top of Mesos while there is no any extra configuration needed and it provides full access to various stateful services (e.g., HDFS/HBase/Kafka). This architecture is used by multiple teams in the company. For instance, the backend application team uses it to aggregate their logs and metrics, and data science team relies on it to do model training while developing the code. In this talk, Shuai Lin will share the experience they learned in details, including caveats they encountered when deploying and maintaining the notebook server.

Speakers
SL

Shuai Lin

Scrapinghub
Shuai Lin is an Infrastructure Engineer in Scrapinghub, a company that provides PaaS for running web crawlers. He graduated from Tsinghua University in 2010 and has been working in the software industry since then. He has been contributing to Apache Mesos and Apache Spark for yea... Read More →


Wednesday June 21, 2017 14:20 - 15:05
Room 309B

15:10

Ocean, A Flexible/Robust Scheduling Platform in Xiaomi - Yin Sun, Xiaomi
Ocean is one of the most important private cloud Infrastructure platform in Xiaomi, which is built by technologies ,like Mesos,Marathon and Docker. This session will describe the best practices of our output from the engineering perspective, and to present how we leverage the advantages of Mesos&Marathon.

Speakers
YS

Yin Sun

Xiaomi Inc.
employer: | 2004 - 2007 engineer, employed by Inspur | 2007 - 2013 Senior architect, employed by Baidu | 2013 - Mentor architect, employed by Xiaomi | | previous speaking experience: | QCon Beijing(2011) | SACC Beijing(2012) | GITC Beijing(2014) | GITC Beijing(2016)


Wednesday June 21, 2017 15:10 - 15:55
Room 309A

15:10

The Best Practices of DevOps over DC/OS - Sam Chen & Ben Lin, Mesosphere APAC
DC/OS is super hot container technology in China. In this talk, Ben and I will introduce how to realize DevOps scenarios with customer references in China. It includes how to architect DevOps, coding, Gitlab, Jenkins for testing, Marathon for deploying, production deployment etc.

In the same time, being able to show the best practices of customer reference cases.

Speakers
SC

Sam Chen

APJ Country Director, Mesosphere APAC
Running business in APJ
BL

Ben Lin

Solutions Architect, Mesosphere
Ben is currently leads all technical initiatives for Mesosphere in APAC. He has presented at MesosCon Asia, VMworld, VMworld Europe, LISA, HotCloud, and other conferences.


Wednesday June 21, 2017 15:10 - 15:55
Room 309B

15:10

Practices of Mesos in 360SearchEngine [C] - Bokan Tan, 360Search
360Search is the second search engine in China, which own more than ten thousands of server. In this presentation, Bokan Tan will introduce the challenge of 360Searchand the practices of Mesos on producation environment,such as service discovery, deploy stateful service on mesos, custom executor and the mesos ecosystem.

Speakers
BT

Bokan Tan

2009-2012 renren-inc (NYSE: RENN) Software Engineer : Develop SNS background system | 2012-2014 sogou-inc Senior Software Engineer : Develop and Maintain HBase 0.89-fb branch in Sogou Search Engine | 2014-2017 360 (NYSE: QIHU) System Architect: Build a PaaS platform for 360Se... Read More →



Wednesday June 21, 2017 15:10 - 15:55
Room 308

15:55

16:30

LXD Containers, Mesos and Infrastructure as a Service - Tom Barber, Meteorite Consulting
Apache Mesos has utilised two main container types, Mesos containers & Docker. Docker containers are the most visible container types in mainstream operations, but they have a major drawback in some use cases, they aren’t an entire system. They have been trimmed down immensely to the bare minimum. LXD offers the ability to run full operating systems & therefore within a Mesos ecosystem. Deploying machine containers, instead of application containers, this allows Mesos to provide Infrastructure as a Service capabilities. In supporting LXD we then extend the software support on Mesos by allowing users to apt-get or yum install software on the nodes rather than being bound to pre built containers. LXD support allows users to deploy any software they choose & leverage existing tooling like Juju, Puppet & Chef over their Mesos resources as Sys Admins would in a traditional deployment.

Speakers
avatar for Tom Barber

Tom Barber

Technical Director, Spicule LTD
Tom Barber is the director of Meteorite BI and Spicule BI. A member of the Apache Software Foundation and regular speaker at ApacheCon, Tom has a passion for simplifying technology. The creator of Saiku Analytics and open source stalwart, when not working for NASA, Tom currently... Read More →


Wednesday June 21, 2017 16:30 - 17:15
Room 309A

16:30

Apache Flink and More [E] - Jörg Schad, Mesosphere & Till Rohrmann, dataArtisans
Apache Mesos allows operators to run distributed applications across an entire datacenter and is attracting ever increasing interest. As much as distributed applications see increased use enabled by Mesos, Mesos also sees increasing use due to a growing ecosystem of well-integrated applications. One of the latest additions to the Mesos family is Apache Flink.

Flink is one of the most popular open source systems for real-time high scale data processing and allows users to deal with low-latency streaming analytical workloads on Mesos.

In this talk, we explain the challenges solved while integrating Flink with Mesos, including how Flink’s distributed architecture can be modeled as a Mesos framework, and how Flink was integrated with Fenzo. Next, we describe how Flink was packaged to easily run on DC/OS.

Speakers
avatar for Till Rohrmann

Till Rohrmann

Engineering lead, data Artisans
Till is a PMC member of Apache Flink and software engineer at dataArtisans. His main work focuses on enhancing Flink’s scalability as a distributed system. Till studied computer science at TU Berlin, TU Munich and École Polytechnique where he specialized in machine learning an... Read More →
JS

Jörg Schad

Software Engineer, Mesosphere
Jörg is a software engineer at Mesosphere in Hamburg. In his previous life he implemented distributed and in memory databases and conducted research in the Hadoop and Cloud area. His speaking experience includes various Meetups, international conferences, and lecture halls.



Wednesday June 21, 2017 16:30 - 17:15
Room 309B

16:30

Mesos-Based Big Data Infrastructure @ Douban [C] - Zhongbo Tian, Douban Inc.
How to build an elastic and efficient platform to support various Big Data and Machine Learning tasks is a challenge for a lot of corporations. In this presentation, Zhongbo Tian will give an overview of the Mesos-based core infrastructure of Douban, and demonstrate how to integrate the platform with state-of-art Big Data/ML technologies.

Speakers
ZT

Zhongbo Tian

Lead of Infrastructure Team, Douban Inc,
Senior System Architect, Lead of Core Infrastructure Team of Douban Inc, with 9 years of expertise in distributed storage system, micro-services architecture and HPC , focusing on high-availability, consistency, scalability and performance. | | Lead the whole team to transfor... Read More →



Wednesday June 21, 2017 16:30 - 17:15
Room 308

17:15

 
Thursday, June 22
 

08:00

Registration
Thursday June 22, 2017 08:00 - 16:30
CNCC

09:00

Keynote: Join the Mesos Community! - Gilbert Song, Distributed Systems Engineer, Mesosphere
There are a lot of Mesos users in China, who have been using Mesos to manage their production clusters for years and have been enjoying the magic that Mesos makes stability and scalability become realistic. However, some of these big Mesos users are still not well known yet. In fact, more involvements into the Mesos community would bring a lot more benefits than one can expect.

In this talk, Gilbert will present great examples (e.g., Uber, Apple, Douban and Dangdang) of how a company gains the most benefits from the Mesos community, while the community also receives great feedbacks from these outstanding users. Then, Gilbert will introduce different ways to get involved with the Mesos community. It is not only limited to writing codes, but also on showing great ideas on design, requesting features, filing issues, and any format of advice on helping our community become insanely great.

Speakers
avatar for Gilbert Song

Gilbert Song

Distributed Systems Engineer, Mesosphere
Gilbert Song, Apache Mesos PMC/Committer, is a Distributed Systems Engineer at Mesosphere. He has been contributing to Mesos for years and mainly focuses on Mesos Containerization. He holds a Master’s degree in Computer Engineering from University of California, Santa Barbara. He... Read More →



Thursday June 22, 2017 09:00 - 09:30
Room 309A

09:30

09:50

Keynote: Building an Ecosystem at JD - Dr. Leslie Liu, Vice President & GM, Enterprise Service Division, JD Cloud
Speakers
avatar for Dr. Leslie Liu

Dr. Leslie Liu

Vice President & GM, Enterprise Service Division, JD Cloud
Dr. Leslie Liu is the vice president of JD Cloud, GM of the enterprise service division. He joined JD Smart in 2014 where he served as VP of JD Smart and GM of smart enterprise division. Dr. Liu is currently leading the enterprise service department with the mission to build the new RaaS (Retail as a service) platform. This new technology and business platform incorporates the capabilities from both... Read More →


Thursday June 22, 2017 09:50 - 10:10
Room 309A

10:10

10:30

Keynote: Big Data at Tsinghua University - Chen Wang, CTO, National Engineering Lab for Big Data Software, Tsinghua University
Academic and system research at National Engineering Lab for Big Data Software of Tsinghua University.

Speakers

Thursday June 22, 2017 10:30 - 11:00
Room 309A

11:00

11:30

Apache Mesos and DC/OS Networking Deep Dive - Jörg Schad, Mesosphere
Apache Mesos and DC/OS allow users to deploy distributed applications and in particular micro-services across a large cluster. Therefore, networking becomes an important aspect especially when trying to provide highly-available applications on top of an unreliable infrastructure.

In this talk we will first present the various challenges around networking for distributed micro-service architectures, including
* Connectivity
* Service Discovery
* Load-balancing
* Isolation

Next, we will discuss different solutions to those problems with the example of DC/OS including:
* Minuteman
* MesosDNS
* Marathon-LB
* Overlay network (including CNI )

Afterwards, we will deep dive into the actual implementation of the different components in order to understand how we can achieve a scalable and flexible networking.

Speakers
JS

Jörg Schad

Software Engineer, Mesosphere
Jörg is a software engineer at Mesosphere in Hamburg. In his previous life he implemented distributed and in memory databases and conducted research in the Hadoop and Cloud area. His speaking experience includes various Meetups, international conferences, and lecture halls.


Thursday June 22, 2017 11:30 - 12:15
Room 309A

11:30

Cloud Containers Practices with Mesos in Ctrip [C] - Feng Cai, Ctrip
Part 1: VM -> Container
We built a private cloud with OpenStack, It manages more than 30k VMs, 3k+ Compute Nodes and Bare Metals. As the business continues to grow, delivery speed is we care about most. In this part, I will show you how we manage container with OpenStack.

Part 2: Openstack -> Mesos.
OpenStack is too heavy for container scheduling. We evaluated Kubernetes, Mesos and Swarm. We choosed Mesos because it has flexible architectures.In this part, I will show you why we choose Mesos and some experiences on Chronos framework.

Part 3: Built our own framework and executor
Chronos/Marathon can't fully meet ourrequirement. We write a framework with Python, a executor with Go. In this part, I'll show you the detail about our framework and executor

Part 4: Windows Container in Mesos
This part, I'll show you the progress we've made on running windows container in Mesos.

Speakers
avatar for Feng Cai

Feng Cai

Cloud Platform Director, Ctrip
I'm Feng Cai, a Director of Cloud platform of Ctrip, is responsible for the R&D of desktop cloud, private cloud and container cloud platform. I used to work in eBay, Sina Corp, SNDA Game and other Internet companies. I have full experience in Linux, large scale distribute system... Read More →



Thursday June 22, 2017 11:30 - 12:15
Room 309B

11:30

How We Became the Largest eCommerce Platform in Southeast Asia via the Mesos Clusters - Haosdent Huang, Mesos Committer
The network environment is very complicated and unstable in Southeast Asia, especially the international bandwidth between different countries. In this presentation, haosdent would introduce the Shopee infrastructure which based on Mesos. He would show how Shopee use Mesos to manage their servers (2000+) in different datacenters across Southeast Asia and use Mesos to boost and scale their services. Haosdent would share Shopee experience about Mesos monitoring, maintenance, service discovery, disaster recovery, load balancing and other interesting parts. It is because of the powerful features of Mesos to help Shopee grow to the largest eCommerce platform in Southeast Asia in one year.  

Speakers
avatar for Haosdent Huang

Haosdent Huang

Shopee
Haosdent (Mesos Committer), Shopee Engineer, Design and implementing Shopee Infrastructure.



Thursday June 22, 2017 11:30 - 12:15
Room 308

12:20

OCI Support in Mesos [C,E] - Qian Zhang, IBM & Jie Yu, Mesosphere
OCI is an open governance project which creates open standards around container formats and runtimes, currently it contains two specifications: the image specification and the runtime specification, the image specification is the focus of how we are going to support OCI in Mesos. Mesos containerizer has already supported Appc image and Docker image, and it is important to support OCI image as well since it is going to be the open industry standard. So in this presentation, Jie and Qian will present the recent work of enhancing Mesos containerizer to support OCI image, in particular, how OCI image is fetched and cached, how OCI image is provisioned to a container's rootfs, how OCI image configuration is consulted to launch a container, and a live demo to show the end-to-end workflow.

Speakers
JY

Jie Yu

Tech Lead, Mesosphere
Jie Yu is a Tech Lead at Mesosphere, Inc, focused on Mesos development. He is a long time Apache Mesos committer and PMC member (since 2014). He leads the development of key container, networking and storage projects at Mesosphere. Before joining Mesosphere, he was a software eng... Read More →
QZ

Qian Zhang

Qian Zhang is a software architect in IBM China, he has more than 10 years working experiences on resource management / scheduling, cloud computing, virtualization / container technologies. Currently he is Apache Mesos committer and PMC member, his contributions covers CNI suppor... Read More →



Thursday June 22, 2017 12:20 - 13:05
Room 309A

12:20

An Introduction to VM Management Framework [C,E] - Xiaohui Liu, UCloud
A mesos framework is the component used to manage mesos tasks. The tasks could be long-running services, big data processing, batch job scheduling etc with different frameworks.
Is it possible to run a virtual machine on mesos? If it is, how to do that? In this presentation, Xiaohui will discuss how to build a framework to run virtual machines on mesos,
describe the design of scheduler and executor of framework in details, and introduce some management experience of network, storage, image. 

Speakers
XL

Xiaohui Liu

Xiaohui,Liu, is a senior software engineer at UCloud. He works mainly on the cluster management.



Thursday June 22, 2017 12:20 - 13:05
Room 309B

12:20

Doing “DropBox” the Cloud Native Way - Jörg Schad, Mesosphere
Cloud Native architecture has slowly become the default way to build robust, scalable applications. How would you rebuild a large scale storage service such as Dropbox (please note Dropbox is just an example here and we have no plans of actually competing with Dropbox), using the Cloud Native architecture?

In this presentation, Joerg and Nitish will discuss the Cloud Native architecture, its advantages, and then explain how to build a scalable, multi-tenant, Dropbox like cloud storage system using modern, containerized applications like Minio as the unstructured data/blob store, MongoDB for structured data, Redis for key value pairs etc, all orchestrated on DC/OS.

The presentation will also include a live demo of the Dropbox equivalent product.

Speakers
JS

Jörg Schad

Software Engineer, Mesosphere
Jörg is a software engineer at Mesosphere in Hamburg. In his previous life he implemented distributed and in memory databases and conducted research in the Hadoop and Cloud area. His speaking experience includes various Meetups, international conferences, and lecture halls.


Thursday June 22, 2017 12:20 - 13:05
Room 308

13:05

Lunch
Thursday June 22, 2017 13:05 - 14:20
Level 3 Foyer

14:20

Executor Deep Dive - Vinod Kone & Gregory Mann, Mesosphere
Historically, Mesos has provided a command executor for running one off tasks by schedulers e.g., Marathon. For launching a group of co-located tasks, the recommended way was to implement the logic using a custom executor. With the support for task groups aka Pods in Mesos, the default executor is the new recommended way for running pods in production.

We believe that everyone using custom executors in their Mesos clusters in production would benefit from learning more about the new features in the default executor. This talk will cover the following specific topics:

- Overview of the Mesos Agent API’s for nested containers
- Authentication and authorization of executors
- Best practices on running sidecar containers and transient tasks
- Health Checks and Probes
- Ops: Best practices for logging & metrics using default executor
- New planned upcoming features on the roadmap



Speakers
avatar for Vinod Kone

Vinod Kone

Distributed Systems Engineer, Mesosphere
Vinod Kone is a committer and PMC member of the Apache Mesos project. He is currently a Tech Lead and Engineering Manager @ Mesosphere. Previously, he was a Tech Lead and Manager of the Mesos team @Twitter. Vinod completed his PhD in Computer Science from UC Santa Barbara.
GM

Gregory Mann

Greg Mann is a distributed systems engineer at Mesosphere and an Apache Mesos contributor. He came to distributed systems by way of high performance computing, performing computational studies of solid state material properties during his time in the physical chemistry PhD progra... Read More →



Thursday June 22, 2017 14:20 - 15:05
Room 309A

14:20

Marathon and Jobs - Today and Tomorrow - Artem Harutyunyan, Director of Engineering, Mesosphere
Marathon and Chronos have been the core and essential frameworks for most Apache Mesos clusters. Marathon ensures long running tasks are running in the datacenter and is commonly used to maintain high availability of other frameworks. Chronos is the fault tolerant cron of the datacenter. Not long ago, Mesosphere introduced Metronome as a replacement for Chronos based on the Marathon code base. There has been significant change in Marathon to leverage the latest features of Apache Mesos, shift the framework to a reliable and stable state and improve its overall performance and scalability.

This session will provide an overview of all the new features, a reasoning about how and why Marathon will grow, and will conclude with a look at the roadmap for Marathon in particular and frameworks launched via Marathon in general.

Speakers
avatar for Artem Harutyunyan

Artem Harutyunyan

Director of Engineering, Mesosphere
Artem Harutyunyan is a Director of Engineering at Mesosphere where he works on Apache Mesos and Marathon projects. Prior to joining Mesosphere he spent several years at Qualys where he led the design and development of a distributed system for storing and analyzing large volumes... Read More →



Thursday June 22, 2017 14:20 - 15:05
Room 309B

14:20

Mesos vs Kubernetes: What We Learned Working With Both From Chinese Customers [C] - Guangya Liu & Yong Feng, IBM
IBM has been working in Apache Mesos community for almost two year and also builds a product named as CfC (Conductor for Container https://hub.docker.com/r/ibmcom/cfc-installer/ ) which is based on Kubernetes and Mesos, the alias for CfC is BlueDock in China. BlueDock is focused on providing container service for customers.

In this talk, we want to talk something as following:
1) Engagement experience and feedback about customer’s attitude towards Mesos and open DC/OS ecosystem compared with Kubernetes.
2) The advantages and disadvantages for Mesos and Kubernetes based container cloud solution
3) The adoption of those different solutions and why customer choose different solutions.
4) How to improve Mesos and Open DC/OS ecosystem to align more with Chinese customer requirement.
5) What IBM is doing to make Mesos and Mesos + Kubernetes awesome for Chinese customers.

Speakers
YF

Yong Feng

Senior Software Architect, IBM Canada Ltd.
Yong Feng is a Senior Software Architect at IBM Platform. He has more than 10 years of experience in design, and implements cluster, grid and cloud computing systems with a focus on scheduling, resource and workload management. He has a deep knowledge on related open-source proje... Read More →
GL

Guangya Liu

Senior Software Architect, IBM
Guang Ya Liu is a Senior Software Architect in IBM CSL (China System Lab) and now focusing on cloud computing, data center operating system and container technology, he is also a Member of IBM Academy of Technology. Starting from 2013, Guang Ya act as an OpenStack Active Contribu... Read More →



Thursday June 22, 2017 14:20 - 15:05
Room 308

15:10

Running Marathon/Mesos on ARM Server [E] - Yuqi Gu & Yibo Cai, ARM
Marathon/Mesos has been well adopted and deployed, in X86 architecture servers. How about running Marathon/Mesos in ARM server? Fortunately, we proved that only small effort is required to port Marathon/Mesos onto ARM AArch64 servers. There are a few small traps and tricks, but the overall porting is straightforward. We would like to present running Marathon/Mesos in ARM servers to manage a cluster on top of which some applications can be deployed and scaled in containers. 

Speakers
avatar for Yibo Cai

Yibo Cai

Staff Software Engineer, ARM
Yibo has been working in IT industry for 17 years. He is Staff Software Engineer of ARM LTD, and he is regarded as an expert of OpenStack on AArch64. He is one of the key contributors of Linaro Developer Cloud and Enterprise Reference Platform (a product ready, open source platfo... Read More →
avatar for Yuqi Gu

Yuqi Gu

Senior Software Engineer, ARM
Yuqi has 8 years working experience in IT industry. He is a Senior Software Engineer of ARM LTD, and his primary focus is on Big Data and Database application enablement and optimization on AArch64 platform, including MySQL, Casandra, Hive, Spark, and Mesos and Marathon.



Thursday June 22, 2017 15:10 - 15:55
Room 309A

15:10

Swan: Another Long Running Task Scheduler [C] - Chunming Xu, Dataman-Cloud
Swan, Another long running task scheduler (Kevin Xu, Dataman-Cloud.inc) - Marathon shouldn't be the only choice for scheduling long running tasks among Mesos community, Swan add diversity to the ecosystem, it borrows idea from Marathon but also added with new thoughts. Application orchestration shouldn't be a standalone component, also service discovery and API gateway should be together with it. Swan help exposing user's services out with single software instead of Marathon + Mesos DNS + Marathon-LB.

Speakers
CX

Chunming Xu

Before joining Dataman-Cloud Beijing two years ago, I have been doing backend development for more then 8 years. Dataman-Cloud Beijing is a startup company which is helping our customers transforming existing IT platforms from old way to new containerized and more flexible way. O... Read More →


Thursday June 22, 2017 15:10 - 15:55
Room 309B

15:10

Building a Private Cloud with Mesos - Zhang Liang
This presentation introduce how to build a private cloud for dangdang.com, details are:
1. Benefits of using Mesos as the infrastructure of the private cloud.
2. Best practices of maintenance high quality private cloud.
3. Briefly describe how to develop a Mesos framework for job cloud.

Speakers
ZL

Zhang Liang

Enthusiasm on clean code, be good at distribute system, high concurrency and big data etc. Focus on architecture design, microservice and container, use java over 10 years. Dedicated open source, lead 2 open sourced projects, elastic-job and sharding-jdbc, there are over thousand... Read More →


Thursday June 22, 2017 15:10 - 15:55
Room 308

15:55

16:30

DMesos - Not only a re-implementation of Mesos [C,E] - Feng Li, EMC
DMesos is a new open source project that based on the D language(http://dlang.org/), which not only aims at offering an alternative to Mesos, but also providing some novel designs.
This talk consists of following aspects:
1) Why we need a new Mesos--the design and architecture of DMesos
2) The current implementation status of DMesos
3) DMesos for the next generation of Cloud Computing and Data Center
4) The latest progress in topic "Mesos on ARM" since MesosCon Asia 2016 and why DMesos is fit for ARM platform

Speakers
FL

Feng Li

Principal Software Engineer, EMC
Focus on mobile software development for 10+ years(has been working in Motorola, Samsung, and Alibaba) in China, | now engaged in Software Define Storage related R&D work at EMC(http://www.emc.com/) in Shanghai. | | 1) 'Mesos on ARM', MesosCon ASIA 2016-Hangzhou, and gave pr... Read More →



Thursday June 22, 2017 16:30 - 17:15
Room 309A

16:30

Visualize data with Zeppelin and Mesos - Jongyoul Lee, ZEPL
One of the success points of data analysis is speed and fast iteration. In a modern way, there are a lot of tools to help to clean and analyze data like Spark. But making your result beautiful is still a problem. Zeppelin reduces the step, moving the result to make a graph or table. In your browser, you can run your Spark job through your own cluster and make a graph directly from your result. Including Spark, you can run JDBC, Flink, and python job, too.

But Zeppelin had a problem. Zeppelin runs just a single machine for a long time. It doesn't fit in production level. Now Zeppelin supports to run its jobs into Mesos Cluster. Even if it makes Zeppelin execute jobs, Zeppelin can share in the benefits of solid fault tolerant features of Mesos.

This talks will show demo from installing, launching Zeppelin to running jobs into Mesos. And it also shows beautiful visualization of Zeppelin.

Speakers
avatar for Jongyoul Lee

Jongyoul Lee

Software Development Engineer, ZEPL
I'm a member of PMC of Apache Zeppelin and works at ZEPL. In Apache Zeppelin, I focus on stabilizing Apache Zeppelin to be used in production level, developing some enterprise features and enhancing Apache Spark/JDBC features. Personally, I'm really interested in distributed and... Read More →


Thursday June 22, 2017 16:30 - 17:15
Room 309B

16:30

Mesos Practice in China Unicom - Jinwu Wei, China Unicom
Linux containers are regarded as the next generation of cloud computing architecture. In this presentation, Dr. Wei will introduce the innovative technology architecture of CU-DCOS1.0(China Unicom DCOS1.0), which is DC/OS+k8sm+unified service gateway.
The user interface of CU-DCOS1.0 will be displayed too.

Two demos will be demonstrated, one is the converged system of China Unicom Unified PaaS with CU-DCOS. Through transforming the old systems to microservices and containerizing applications, the management of the converged China Unicom Unified PaaS system became more effective and flexible.

The other demo resource sharing between big data clusters shows how to provides big data service on CU-DCOS1.0. CU-DCOS1.0 provides support for Hadoop, Hive and Spark through the integration of DCOS and Myriad framework. And it solves the problem of load balance for traditional big data clusters.

Speakers
JW

Jinwu Wei

China Unicom
China Unicom is one of the three major telecoms in China, which has a wide influence on the telecom industry.The ongoing project is China Unicom DCOS2.0.


Thursday June 22, 2017 16:30 - 17:15
Room 308