Cloud
Computing
Lecture 2
Datacenter internals
Amazon Getting Started
Dan Amiga
[email protected]
Dan Amiga – IDC Cloud Computing 2012
Problems
• Time
• Cost
• Scale
• Storage Size
• Availability
• Connectivity
• Security
• How-to-do-it-right
Dan Amiga – IDC Cloud Computing 2012
Why Now?
• It’s possible
• Sometimes It’s mandatory
• “Web Space Race”
– Build extremely large datacenters (10,000/0’s)
– Driven by growth (more users, more data)
• Operations & Infrastructure expertise
• Broadband got better
Dan Amiga – IDC Cloud Computing 2012
Cloud Principals
Pay only for what you use
Ability to scale up and scale down
Dan Amiga – IDC Cloud Computing 2012
Cloud Services
Packaged Infrastructure Platform Software
(as a Service)
Software (as a Service) (as a Service)
You manage
Applications Applications Applications Applications
You manage
Data Data Data Data
Runtime Runtime Runtime Runtime
Managed by vendor
Middleware Middleware Middleware Middleware
You manage
Managed by vendor
O/S O/S O/S O/S
Managed by vendor
Virtualization Virtualization Virtualization Virtualization
Servers Servers Servers Servers
Storage Storage Storage Storage
Networking Networking Networking Networking
State of Cloud Computing
• Perceptions > Types
– “The end of software” • Public
– On-demand infrastructure • Private
– Cheaper and better • Internal
• Reality • External
– Hybrid world; not “all-or-nothing” • Hybrid
– Leverage existing IT skills and
investments > Categories
– Seamless user experiences • SaaS
– Evolutionary; not revolutionary • PaaS
• Drivers • IaaS
– Ease-of-use, convenience
– Product effectiveness
– Simplify IT, reduce costs
Dan Amiga – IDC Cloud Computing 2012
Amazon Prices March 2012
Questions:
1. Is this cheap or expensive? Why?
2. Will windows be cheaper? Why?
3. Why are the memory numbers are floats?
4. How can I make this cheaper?
Dan Amiga – IDC Cloud Computing 2012
Spot Instances – bid!
Dan Amiga – IDC Cloud Computing 2012
Dan Amiga – IDC Cloud Computing 2012
Inside a datacenter
Dan Amiga – IDC Cloud Computing 2012
Dan Amiga – IDC Cloud Computing 2012
Dan Amiga – IDC Cloud Computing 2012
Dan Amiga – IDC Cloud Computing 2012
Virtualization
• Many virtualized X on top of real X
• Types
– Full (VMWare, Intel VT)
– Hardware assist
– Partial
– Paravirtualization (Xen)
Dan Amiga – IDC Cloud Computing 2012
Virtual Machine Monitors
• VMM is just software
– Applications not modified (so we just deploy our code..)
– Do you see any issues with sharing?
VM0 App0 VM1 App1 VMn Appn
...
Guest OS0 Guest OS1 Guest OSn
Virtual Machine Monitor (VMM)
Platform HW
Memory Processor/CS I/O Devices
Dan Amiga – IDC Cloud Computing 2012
Virtualization Capabilities
Workload Isolation Workload Consolidation
App1 App2 App1 App2 App1 App2 App1 App2
OS OS OS OS1 OS2 OS1 OS2
HW VMM HW1 HW2 VMM
HW HW
Workload Migration Workload Embedding
App App App App
OS OS OS1 OS2
VMM VMM VMM VMM VMM
HW1 HW2 HW1 HW2 HW
Virtualization has powerful capabilities
(Slide taken from “Understanding Intel VT”)
Dan Amiga – IDC Cloud Computing 2012
Virtualization Challenges
• Functionality
• Overhead
• Speed & Performance
• Security (Resource Isolation, e.g. UDP)
Dan Amiga – IDC Cloud Computing 2012
XEN Hypervisor (Amazon)
• Paravirtualization
– Requires modifying the underlying OS to
overcome and simplify virtualization challenges
– One of the reasons not all OS’s are applicable for
EC2
– BVT Scheduling
• Virtualization is for
– CPU
– Memory Increasing Privilege Level
Ring 0
– IO Ring 1
Ring 2
Dan Amiga – IDC Cloud Computing 2012 Ring 3
Amazon Xen benchmarks
• The Impact of Virtualization on Network Performance of Amazon EC2 Data Center (Guohui
Wang T. S. Eugene Ng Dept. of Computer Science, Rice University)
Dan Amiga – IDC Cloud Computing 2012
Amazon CPU Utilization
• Average 7.3% with possible CPU cap
allocation
Dan Amiga – IDC Cloud Computing 2012
Warehouse scale computers
• Massive Scale - ~10K-100K servers
• Homogenous hw/sw
• Lots of attention to energy savings and efficiency
• Design goals
– Needs to be highly available (or it hits the news)
– Doesn’t solve all problems
– Competitive (prices always go down)
Dan Amiga – IDC Cloud Computing 2012
No “welcome to Google” sign
• What is the unit of modularity?
• Sometimes no building..
• http://www.youtube.com/watch?v=zRwPSFpLX8I
Dan Amiga – IDC Cloud Computing 2012
Dan Amiga – IDC Cloud Computing 2012
Dan Amiga – IDC Cloud Computing 2012
Power Usage Effectiveness (PUE)
• Less conversion from of DC/AC is a must
• How efficient you deliver power and cooling
to your servers; 1 is optimal
Dan Amiga – IDC Cloud Computing 2012
Facebook data centers
• Is Facebook a cloud provider?
• Customized racks (3 columns)
• Next to a rack there is
– DC power strip (in case of utility failure)
– DC UPS
– AC power strip
• 30 Servers per column, 1.5 U per server (why?)
• https://www.facebook.com/note.php?note_id=1
0150144039563920
• http://opencompute.org/specs/
• PUE = 1.07
Dan Amiga – IDC Cloud Computing 2012
Server Drill Down
• Servers
• Racks (U)
– Rack format
– 40-80 servers
– Communication via
In-rack-switch
• Clusters of racks
– Communication through
Cluster switch (expensive)
Dan Amiga – IDC Cloud Computing 2012
Google Server
Dan Amiga – IDC Cloud Computing 2012
Amazon Data Center
• http://huanliu.wordpress.com/2012/03/13/amazon-data-center-size/
• Each EC2 instance is assigned an internal IP address in the form of 10.x.x.x.
• Each server rack is assigned a 10.x.x.x/22 IP address range, i.e., all virtual
machines running on that server rack will have the same 22 bits IP prefix.
• A 10.x.x.x/22 IP address range has 1024 IP addresses, but the first 256 are
reserved for DOM0 virtual machines (system management virtual machine in
XEN), and only the last 768 are used for customers’ instances.
• Within the first 256 addresses, two at address 10.x.x.2 and 10.x.x.3 are reserved
for routers on the rack. These two routers are arranged in a load balanced and
fault-tolerant configuration to route traffic in and out of the rack. I verified that
the uplink capacity from 10.x.x.2 and 10.x.x.3 are roughly 2 Gbps total, further
suggesting that they are routers each with a 1Gbps uplink.
Dan Amiga – IDC Cloud Computing 2012
Amazon Data Center Size (03/13/2012)
data center\size # of server racks # of blade servers
US East (Virginia) 5,030 321,920
US West (Oregon) 41 2,624
US West (N.
630 40,320
California)
EU West (Ireland) 814 52,096
AP Northeast
314 20,096
(Japan)
AP Southeast
246 15,744
(Singapore)
SA East (Sao
25 1,600
Paulo)
Total 7,100 454,400
Dan Amiga – IDC Cloud Computing 2012
Memory/Storage
• Where is the Lie? What’s on the top?
Dan Amiga – IDC Cloud Computing 2012
Dan Amiga – IDC Cloud Computing 2012
Hard drives
Dan Amiga – IDC Cloud Computing 2012
SSD Hard drive
• SSD FLASH vs SSD DRAM
• Half to third of power
• Reliability and lifetime
• Software Controller
– ECC, Wear leveling, Caching, GC,
• Price per GB will fall below $1 for 1GB
• Dynamo DB
Dan Amiga – IDC Cloud Computing 2012
Principal of locality + WSC
Latency vs Throughput Architecture consideration
Local Rack Array
Rack 1 30
Servers 1 80 2400
Cores 8 640 19200
DRAM 16 1280 38400
DISK 4000 320000 9600000
DRAM Latency 0.1 100 300
Disk Latency 10,000 11,000 12,000
DRAM Bandwidth 20,000 100 10
Disk Bandwidth 200 100 10
Dan Amiga – IDC Cloud Computing 2012
Dan Amiga – IDC Cloud Computing 2012
