Slacker: Fast Distribution with Lazy Docker Containers

Containerized applications are becoming increasingly popular, but unfortunately, current container-deployment methods are very slow. We develop a new container benchmark, HelloBench, to evaluate the startup times of 57 different containerized applications. We use HelloBench to analyze workloads in detail, studying the block I/O patterns exhibited during startup and compressibility of container images. Our analysis shows that pulling packages accounts for 76% of container start time, but only 6.4% of that data is read. We use this and other findings to guide the design of Slacker, a new Docker storage driver optimized for fast container startup. Slacker is based on centralized storage that is shared between all Docker workers and registries. Workers quickly provision container storage using backend clones and minimize startup latency by lazily fetching container data. Slacker speeds up the median container development cycle by 20× and deployment cycle by 5×.

[1]  Andrea C. Arpaci-Dusseau,et al.  Physical Disentanglement in a Container-Based File System , 2014, OSDI.

[2]  Jon Howell,et al.  Handling Flash Crowds from Your Garage , 2008, USENIX Annual Technical Conference.

[3]  Gregory R. Ganger,et al.  Argon: Performance Insulation for Shared Storage Servers , 2007, FAST.

[4]  Armando Solar-Lezama,et al.  Towards optimization-safe systems: analyzing the impact of undefined behavior , 2013, SOSP.

[5]  Hitesh Ballani,et al.  End-to-end Performance Isolation Through Virtual Datacenters , 2014, OSDI.

[6]  Scott Devine,et al.  Disco: running commodity operating systems on scalable multiprocessors , 1997, TOCS.

[7]  Amin Vahdat,et al.  Enforcing Performance Isolation Across Virtual Machines in Xen , 2006, Middleware.

[8]  Dirk Merkel,et al.  Docker: lightweight Linux containers for consistent development and deployment , 2014 .

[9]  Aman Kansal,et al.  Q-clouds: managing performance interference effects for QoS-aware clouds , 2010, EuroSys '10.

[10]  Anees Shaikh,et al.  Performance Isolation and Fairness for Multi-Tenant Cloud Storage , 2012, OSDI.

[11]  Kang G. Shin,et al.  Maestro: quality-of-service in large disk arrays , 2011, ICAC '11.

[12]  Andrea C. Arpaci-Dusseau,et al.  Split-level I/O scheduling , 2015, SOSP.

[13]  Randy H. Katz,et al.  Mesos: A Platform for Fine-Grained Resource Sharing in the Data Center , 2011, NSDI.

[14]  Remzi H. Arpaci-Dusseau.  Operating Systems: Three Easy Pieces , 2015, login Usenix Mag..

[15]  John Pescatore,et al.  Nimda Worm Shows You Can't Always Patch Fast Enough , 2005 .

[16]  Anoop Gupta,et al.  Performance isolation: sharing and isolation in shared-memory multiprocessors , 1998, ASPLOS VIII.

[17]  Carl A. Waldspurger,et al.  Memory resource management in VMware ESX server , 2002, OSDI '02.

[18]  Gabriel Antoniu,et al.  Going back and forth: efficient multideployment and multisnapshotting on clouds , 2011, HPDC '11.

[19]  Abhishek Verma,et al.  Large-scale cluster management at Google with Borg , 2015, EuroSys.

[20]  Monica S. Lam,et al.  Optimizing the migration of virtual computers , 2002, OPSR.

[21]  Patrick E. O'Neil,et al.  The log-structured merge-tree (LSM-tree) , 1996, Acta Informatica.

[22]  Xiao Zhang,et al.  CPI2: CPU performance isolation for shared compute clusters , 2013, EuroSys '13.

[23]  Andrea C. Arpaci-Dusseau,et al.  ANViL: Advanced Virtualization for Modern Non-Volatile Memory Devices , 2015, FAST.

[24]  Ole Agesen,et al.  A comparison of software and hardware techniques for x86 virtualization , 2006, ASPLOS XII.

[25]  Song Jiang,et al.  TotalCOW: Unleash the Power of Copy-On-Write for Thin-provisioned Containers , 2015, APSys.

[26]  Jun Zhu,et al.  Twinkle: A fast resource provisioning mechanism for internet services , 2011, 2011 Proceedings IEEE INFOCOM.

[27]  Mike Hibler,et al.  USENIX Association Proceedings of the General Track : 2003 USENIX Annual , 2003 .

[28]  Sebastien Goasguen,et al.  Image Distribution Mechanisms in Large Scale Cloud Providers , 2010, 2010 IEEE Second International Conference on Cloud Computing Technology and Science.

[29]  Eyal de Lara,et al.  SnowFlock: rapid virtual machine cloning for cloud computing , 2009, EuroSys '09.

[30]  Michael D. Ernst,et al.  An experimental evaluation of continuous testing during development , 2004, ISSTA '04.