• No results found

Cloud Computing. Dipl.-Wirt.-Inform. Robert Neumann

N/A
N/A
Protected

Academic year: 2021

Share "Cloud Computing. Dipl.-Wirt.-Inform. Robert Neumann"

Copied!
42
0
0

Loading.... (view fulltext now)

Full text

(1)

Cloud Computing

(2)

Pre-Cloud Provisioning

Load Forecast Real Load Provisioned IT Capacity Overbuy Underbuy Fixed Cost for Capacity IT Capacity Investment Hurdle

(3)

145 Cloud Computing with Azure

Cloud Elasticity

Zeit IT Kapazitäten Load Precast Real Load Provisioned IT Capacity Lower Initial Invest Reduction of Overbuy Elimination of Underbuy Reduction of Capacity with Declining Load

(4)

Cloud Characteristics

-No up-front investment into capital

-Pay as you go

-Alignment of capacity with demand

1. XaaS 2. Virtuali-zation 3. Multi-tenancy 4. Business Model Cloud

Cloud Components

(5)

147 Cloud Computing with Azure

1. XaaS

Software as a Service (SaaS) Platform as a Service (PaaS) Infrastructure as a Service (IaaS)

Desktop as a Service (DaaS) Employee as a Service (EaaS) Gaming as a Service (GaaS)

(6)

PaaS

Service Provider (SP) responsible for

- Hosting software - Disaster recovery - Fail-over

(7)

149 Cloud Computing with Azure

SaaS (e.g., Salesforce)

Advantages

- No up-front investment - Transparent IT cost

- Fast implementation

- Process complexity reduction - Procses mobility

Disadvantages

- Dependency on service provider - Slow data transfer rates

- Reduced adaptability

(8)

PaaS (e.g., Azure, App Engine)

Advantages

- Computing platform/solution stack - Facilitates deployment of applications - (.Net, JVM)

Disadvantages

(9)

151 Cloud Computing with Azure

IaaS (e.g., EC2, S3)

Advantages

- Platform virtualization environment - Compute resources as a service - (VMs, Networks, Storage, etc.)

Disadvantages

(10)

2. Virtualization

-Goal:

- Enhance utilization of computing resources - Google server utilization ~ 60%

• Partial Virtualization

• Paravirtualization

• Full Virtualization

(11)

153 Cloud Computing with Azure

Partial Virtualization

Characteristics

- Simulation of multiple instances of hardware - In particular: address spaces

- To have independent address spaces for applications - “VMs“ do not host entire OS‘s!

(12)

Paravirtualization

Characteristics:

- VM offers special API that can only be used by modified guest OS‘s

- “hypercalls“ to hyerpvisor

(13)

155 Cloud Computing with Azure

Full Virtualization

Characteristics

- Virtual machine simulates enough hardware to run unmodified guest OS in isolation

- Originated by IBM‘s CP-40, CP-67 in 1966

(14)

Hypervisors/Virtual Machine Monitor (VMM)

Type1 (Bare metal)

- Run directly on host hardware

Type2 (hosted)

- Run on conventional operatings systems

Products

(15)

157 Cloud Computing with Azure

Ring priviliges

Windows XP:

-Kernel Mode (Ring0) -User Mode (Ring3)

(16)

Ring privileges

Hypervisors

Guest has no direct Access to hardware

(17)

159 Cloud Computing with Azure

Virtualization Performance

1. Ring deprivileging

 Running guest OS at ring higher 0

2. Virtualization of protected mode

 Binary translation  Shadow Page Tables  I/O device emulation

(18)

Binary translation

-Rewrite certain instructions that would fail or behave

differently when executed above ring 0

- Segment executable into basic blocks

- Translate basic blocks to and run on hardware

-Static

- Translate into basic blocks without running executable - Difficult to find all basic blocks

-Dynamic

- Translate basic blocks as they are discovered while running executable

(19)

161 Cloud Computing with Azure

Shadow page tables

- Duplicating Memory Management Unit (MMU)

- Contain the guest-virtual to host-physical address mapping

-Principle

- Denying guest direct access to MMU by: - Trapping access atempts

- Emulating them in software

src: http://www.anandtech.com/ show/2480/10

(20)

Translation Lookaside Buffer (TLB)

-Part of the MMU

-Address translation is compute-intensive

-TLB buffers logical to physical address translation

(21)

163 Cloud Computing with Azure

Nested Paging (AMD), EPT (Intel)

-Context switch (VMExit/VMEntry) very heavy!

- Flushes TLB

- VM-specific Address Space Identifier (ASID)

src: http://www.anandtech.com/ show/2480/10

TLB caches both:

-Virtual memory (Guest) to physical memory (Guest) -Physical memory (Guest) to Physical memory (Hardware)

Perf. Boost up to 23% (AMD)

(22)

I/O device emulation

-Unsupported devices on the guest must be emulated by a

device emulator that runs in the host

(23)

165 Cloud Computing with Azure

Popek & Goldberg Virtualization Requirements

-Equivalence

- VM running on VMM must achieve same behavior as native

-Resource control

- VMM must be in complete control of virtualized resources

-Efficiency

- Majority of instructions must be executed without VMM intervention

(24)

Hardware Assisted Full Virtualization (Hardware Assist)

-Full virtualization using hardware capabilities

- Intel VT-x, AMD V

- Added to X86 in 2006

(25)

167 Cloud Computing with Azure

Cloud Scheduling

Cloud Optimization Problem:

1. Avoid wasting resources

 Through under-utilization

2. Avoid lengthy response times

(26)

Cloud Scheduling

Assumptions:

- PM denotes a physical machine, n: |{PM}| - VM denotes a virtual machine, m: |{VM}| - VMPM: m1

- It is desirable to:

- Avoid under-utilization

 Have as many PMs online as necessary

- Avoid unstable performance

 Have just as many VMs assigned to one PM as

(27)

169 Cloud Computing with Azure

Cloud Scheduling

Cloud Frames

- Number of PMs might change between frames as: - Number of VMs has changed

- Problematic VMs require migration

S: „Schedule“

Requires reallocation of VMs and PMs

S(m, n): mn Sopt

Resource allocation problem: NP-hard! (Vector Bin Packing Problem)

(28)

Cloud Scheduling

Multiple Criteria Decision Analysis (MCDA)

1. Feasible

• In terms of number of migrations

• Number of migrations: |{Migrations}|Max

2. Computationally efficient

• In terms of terminates within reasonable amount of time • Schedule compute time: tComputeMax

FCOpt: „Feasibility-ComputationalEfficiency Optimization“

FCOpt(tComputeMax, |{Migrations}|Max): SSFCOpt (S(m, n)opt)FCOpt

(29)

171 Cloud Computing with Azure

MCDA

1. Choosing a VM to migrate from the list of VMs that run on the problematic PM

2. Choosing a PM to migrate the chosen VM to 3. Migrating the chosen VM to the chosen PM

(30)

Cloud Scheduling Variations

Decentral

- Simple distributed method (SDM)

- Migrate first VM on problematic PM to first available PM - Multiple Criteria Decision Analysis (MCDA)

Central (requires arbiter)

- First Fit (FF)

- First Fit Decreasing (FFD)

(31)

173 Cloud Computing with Azure

(32)
(33)

175 Cloud Computing with Azure

Azure Service Platform

VMs:

Windows Server 2008 R2

VMM:

Windows Azure

(34)

The Azure Platform

1. Compute

1. Web role 2. Worker role

2. Storage

1. Table storage 2. Blob storage 3. Queue storage

3. Fabric

(35)

177 Cloud Computing with Azure

Azure App Fabric

-Access Control Service

- Identify users and grant access

-Service Bus

- Connection via communication and messaging protocols

-Caching

- Distributed in-memory cache for Windows and SQL Azure

-Integration

- Several components of Biztalk

-Composite App

(36)

Azure App Fabric

(37)

179 Cloud Computing with Azure

3. Multitenancy

Definition:

- Single instance of software runs on a server serving multiple clients (tenants)

- Applications are able to virtually partition tenant data and logic

-Multitenancy economies

- Cost savings (per-tenant, on-demand, as-you-go) - Harder to develop due to higher complexity

- Eased release management

-Opposite of multi-instance architecture!

(38)

4. Cloud Business Models

State of the Art:

- RAM + CPU: per compute hour - HDD: per Gbyte/month

- Network: per Gbyte in or out/month

Cloud Spot Markets:

- Procure compute resources in auctions - (e.g., spotcloud.com)

(39)

181 Cloud Computing with Azure

Service Level Agreements (SLAs)

-Part of a service contract

-Measurable details:

- MTBF - Availability - Throughput - Response time - etc.

(40)

Availability

Level of Availability Downtime per year

99% 87.6h

99.9% 8.76h

99.99% 52.56m

99.999% 5.26m

99.9999% 31.56s

(41)

183 Cloud Computing with Azure

Balancing 9s with cost

Industry IT Service Cost/min IT Cost/y

Financial Brokerage operations

$107,500 $58,050,000 Financial Credit card

authorization

$43,333 $23,399,820 Retail Home shopping $1,883 $1,016,820 Retail Catalog sales $1,500 $810,000 Transportation Airline reservation $1,483 $800,820

Finance ATM fees $241 $130,140

(42)

Literature

-

http://www.itsmsolutions.com/newsletters/DITYvol2iss47.ht

m

-

http://www.cs.rutgers.edu/~rmartin/teaching/spring06/cs55

3/papers/002.pdf

-

https://dspace.library.uvic.ca:8443/bitstream/handle/1828/2

420/cloud2010.pdf?sequence=1

References

Related documents

the trachea, heart, cranial vena cava, thoracic aorta, caudal vena cava, right and left principal 27.. bronchus, right and left caudal lobar bronchus and the concomitant branch of

The lungs are connected to the right ventricle of the heart by the pulmonary trunk and its branches, and to the left atrium of the heart by the pulmonary veins.. The lungs

The amount of small business credit extended by lenders to The approval rate of small business credit/loan applications to The aggregate amount of credit requested by small

The survey found UAE female nationals are significantly more likely than their male counterparts to want to work for organisations related to the banking & finance sector.

Now, through Peoples Bank, we bring our tailored mortgage solutions to provide our customers and strategic partners with a range of mortgage products to meet a diverse range

As seen in the previous chapters, the body and its interactions with the surroundings enables learning in a multimodal fashion: it is fundamental for learning even when

It is interesting to learn how Southeast Asian fighter organization founded and grew in Syria and Iraq, poses direct threats to government and the people in the

In the Chrome, ap- proximately all requests sent by browser are accepted from server (code 200). The status “Moved temporarily” means temporary transition of one web address to