Albert Greenberg
James Hamilton
David A. Maltz
Abstract
The data centers used to create cloud services represent a significant investment in capital outlay and ongoing costs. Accordingly, we first examine the costs of cloud service data centers today. The cost breakdown reveals the importance of optimizing work completed per dollar invested. Unfortunately, the resources inside the data centers often operate at low utilization due to resource stranding and fragmentation. To attack this first problem, we propose (1) increasing network agility, and (2) providing appropriate incentives to shape resource consumption. Second, we note that cloud service providers are building out geo-distributed networks of data centers. Geo-diversity lowers latency to users and increases reliability in the presence of an outage taking out an entire site. However, without appropriate design and management, these geo-diverse data center networks can raise the cost of providing service. Moreover, leveraging geo-diversity requires services be designed to benefit from it. To attack this problem, we propose (1) joint optimization of network and data center resources, and (2) new systems and mechanisms for geo-distributing state.
- Amazon Web Services. URL http://aws.amazon.com.Google Scholar
- Engineering @ Facebook's Notes: Scaling Out. URL http://www.facebook.com/notes.php?id=9445547199.Google Scholar
- Google app engine. URL http://code.google.com/appengine/.Google Scholar
- Google docs and spreadsheets. URL http://docs.google.com.Google Scholar
- Microsoft office live. http://office.live.com.Google Scholar
- The Green Grid. URL http://www.thegreengrid.org.Google Scholar
- The Uptime Institute. URL http://uptimeinstitute.org.Google Scholar
- Windows Azure. URL http://www.microsoft.com/azure/.Google Scholar
- Yahoo! Mail. URL http://mail.yahoo.com.Google Scholar
- M. Al-Fares, A. Loukissas, and A. Vahdat. A scalable, commodity data center network architecture. In SIGCOMM, 2008. Google ScholarDigital Library
- L. A. Barroso and U. Hlzle. The case for energy-proportional computing. IEEE Computer, 40, 2007. Google ScholarDigital Library
- A. Brown and D. A. Patterson. Embracing Failure: A Case for Recovery-Oriented Computing (ROC). In High Performance Transaction Processing Symposium, 2001.Google Scholar
- K. Church, J. Hamilton, and A. Greenberg. On delivering embarassingly distributed cloud services. In Hotnets VII, October 2008.Google Scholar
- Cisco. Data center ethernet. http://www.cisco.com/en/US/-netsol/ns783/networking solutions package.html.Google Scholar
- Cisco systems: Data center: Load balancing data center services, 2004.Google Scholar
- A. Greenberg, P. Lahiri, D. A. Maltz, P. Patel, and S. Sengupta. Towards a next generation data center architecture: Scalability and commoditization. In PRESTO Workshop at SIGCOMM, 2008. Google ScholarDigital Library
- C. Guo, H. Wu, K. Tan, L. Shiy, Y. Zhang, and S. Luz. Dcell: A scalable and fault-tolerant network structure for data centers. In SIGCOMM, 2008. Google ScholarDigital Library
- J. Hamilton. Architecture for modular data centers. In Third Biemnial Conference on Innovative Data Systems, 2007.Google Scholar
- IEEE802.1Q. IEEE Standard for Local and Metropolitan Area Networks: Virtual Bridged Local Area Networks, 2005.Google Scholar
- M. Isard. Autopilot: Automatic data center management. Operating Systems Review, 41(2), 2007. Google ScholarDigital Library
- Z. Kerravala. Configuration management delivers business resiliency. The Yankee Group, Nov 2002.Google Scholar
- R. Kohavi, R. M. Henne, and D. Sommerfield. Practical Guide to Controlled Experiments on the Web: Listen to Your Customers not to the HiPPO. KDD, 2007. Google ScholarDigital Library
- C. Kopparapu. Load Balancing Servers, Firewalls, and Caches. John Wisely & Sons Inc., 2002. Google ScholarDigital Library
- E. R. Hinden. Virtual router redundancy protocol (VRRP). RFC 3768, 2004. Google ScholarDigital Library
- W. Enck et al. Configuration Management at Massive Scale: System Design and Experience. IEEE JSAC -- Network Infrastructure Configuration, 2008. Google ScholarDigital Library
Index Terms
- The cost of a cloud: research problems in data center networks
Recommendations
Challenges in cloud scale data centers
SIGMETRICS '13: Proceedings of the ACM SIGMETRICS/international conference on Measurement and modeling of computer systemsData centers are fascinating places, where the massive scale required to deliver on-line services like web search and cloud hosting turns minor issues into major challenges that must be addressed in the design of the physical infrastructure and the ...
Challenges in cloud scale data centers
Performance evaluation reviewData centers are fascinating places, where the massive scale required to deliver on-line services like web search and cloud hosting turns minor issues into major challenges that must be addressed in the design of the physical infrastructure and the ...
Networking the Cloud
ICDCS '09: Proceedings of the 2009 29th IEEE International Conference on Distributed Computing SystemsThe data centers used to create cloud services represent a significant investment in capital outlay and ongoing costs. We examine the costs of cloud service data centers today, and discuss challenges in optimizing work completed per dollar invested. To ...
Comments