際際滷

際際滷Share a Scribd company logo

Documento11propuesta

Download as PPT, PDF
O
Oscar Trenado

The document proposes a software called ESSCCAPE that uses cloud computing to reduce energy use and costs in data centers. It aims to dynamically adjust hardware resources based on workload demands, allowing underutilized servers and components to be powered off when not needed. This approach could help data centers operate more efficiently and avoid overprovisioning hardware to meet peak demands.

1 of 12
Download to read offline
Energy Saving Softwarebased onCloud Computing for Adjustable Processing Environments ( ESSCCAPE ) The Green Cloud
Reduced energy use and therefore reduced energy expenditure. Direct measurable cost savings and value delivery. Carbon emission reductions. Anticipate future green regulations. Improved corporate branding/image. Competitive differentiator: become preferred Green IT partner. Improved management capabilities for data center operation. Introduction Why Green IT ?
Introduction Who is thinking in green ?
Dynamically adding or removing hardware resources could help to stop the energy misuse. A generic approach is needed to control server hardware usage among different tasks. A scalable approach is needed: not only increasing the number of nodes, but also resizing the virtualized hardware resources and/or outsourcing the computational needs (i.e. Amazon EC2). Security : virtualization isolates data and computing resources from potential attacks. Incremental Scalability . Agility/Flexibility . Service-oriented . Public clouds : pay-per-use model. Introduction Why Cloud? From Buzzword to a Real Alternative
Introduction Some Cloud Platforms, Providers and Enablers What do the following companies all have in common: Amazon, Microsoft, Google, EMC, VMware, IBM, Sun, Dell, Akamai, SalesForce.com, NetSuite, and Activision. Answer: Cloud Computing. Merrill Lynch analysts reckon that by 2011 the volume of cloud computing market opportunity will amount to $160BN, including $95BN in business and productivity apps (e-mail, office, CRM, etc.) and $65BN in online advertising. Merrill Lynch.
Nowadays, Green IT is a hot topic for the industry. The problem is that... Data Centers are scaled in order to cope with computational peak workloads. Data Centers are overdimensioned . Server processors are idle most of the time. Data Center energy saving is a huge opportunity for companies and there are a great amount of initiatives to maximize the energy efficiency, thus reducing operational and infrastructure costs. To achieve the goal of Energy Efficiency we need: A well-dimensioned Data Center which uses the resources/components only when needed . The peak workloads are covered by outsorcing the computational needs. Project Overview Our point of view It is not a green approach ... ... A green approach is ...
Virtualization technologies and Cloud computing play a key role in this context. Our proposal aims at: developing an integrated solution to monitor and control the energy consumption of all the Cloud Nodes. taking advantage of all the available information: energy consumption ratings of all the components, current wokload in each Cloud Node, data rates, computational resources location, etc. dynamically adapting the active hardware to the needs in terms of computing and communication resources which are required by the running services. The Green Cloud Project Overview Our point of view
In four steps: obtaining information about the Cloud Nodes: components, workloads, computing resources locations, etc. generating policies to improve the energy consumption, using all the available information. applying this policies in the Cloud Nodes. determining whether the components of the Cloud fit the needs for the future workload or an increase / reduction of hardware is advisable. Green Scope: Generic issues not aiming at reducing the energy waste which are common to Cloud Computing infrastructures are not the focus of this proposal Project Overview Our point of view
Our solution covers the green area . Nevertheless, a cloud is a cloud (i.e. for Amazon its cloud is a private one) so this approach is applicable to both sides (with some special features). The Company Staffs machines are part of the problem and part of the solution too. Project Overview Public or Private? Both Private Cloud (DataCenter) Public Cloud, i.e. company staffs machines
Resources Layer ... Project Overview Architecture. Our point of view ESS = Energy Saving System NCD = Network Card Driver HM = Hardware Manager PM = Processes Manager PA = Power Analizer (optional) NIAC HM NIAC NIAC Measurements DataBase PM HM PM HM PM PA Service Service Service Service Service Service Layer Control Layer Execution DataBase Third Parties like Amazon EC2 Deployment Manager Learning Manager Learnt Policies NCD NCD NCD SNMP Agent SNMP Agent SNMP Agent ESS ESS ESS Processor Layer Processor Allocation Manager Cloud Controller Learnt Policies V.M. V.M. V.M. V.M. V.M.
Oscar Trenado Mu単oz Innovation Manager Tlf: +34 983 54 99 11 +34 606 10 73 50 email: [email_address] Miguel ngel Pe単alvo Bedia Innovation Consultant Tlf: +34 983 54 99 11 e-mail: [email_address] Contact us!
Energy Saving Softwarebased onCloud Computing for Adjustable Processing Environments ( ESSCCAPE ) The Green Cloud

More Related Content

Documento11propuesta

  • 1. Energy Saving Softwarebased onCloud Computing for Adjustable Processing Environments ( ESSCCAPE ) The Green Cloud
  • 2. Reduced energy use and therefore reduced energy expenditure. Direct measurable cost savings and value delivery. Carbon emission reductions. Anticipate future green regulations. Improved corporate branding/image. Competitive differentiator: become preferred Green IT partner. Improved management capabilities for data center operation. Introduction Why Green IT ?
  • 3. Introduction Who is thinking in green ?
  • 4. Dynamically adding or removing hardware resources could help to stop the energy misuse. A generic approach is needed to control server hardware usage among different tasks. A scalable approach is needed: not only increasing the number of nodes, but also resizing the virtualized hardware resources and/or outsourcing the computational needs (i.e. Amazon EC2). Security : virtualization isolates data and computing resources from potential attacks. Incremental Scalability . Agility/Flexibility . Service-oriented . Public clouds : pay-per-use model. Introduction Why Cloud? From Buzzword to a Real Alternative
  • 5. Introduction Some Cloud Platforms, Providers and Enablers What do the following companies all have in common: Amazon, Microsoft, Google, EMC, VMware, IBM, Sun, Dell, Akamai, SalesForce.com, NetSuite, and Activision. Answer: Cloud Computing. Merrill Lynch analysts reckon that by 2011 the volume of cloud computing market opportunity will amount to $160BN, including $95BN in business and productivity apps (e-mail, office, CRM, etc.) and $65BN in online advertising. Merrill Lynch.
  • 6. Nowadays, Green IT is a hot topic for the industry. The problem is that... Data Centers are scaled in order to cope with computational peak workloads. Data Centers are overdimensioned . Server processors are idle most of the time. Data Center energy saving is a huge opportunity for companies and there are a great amount of initiatives to maximize the energy efficiency, thus reducing operational and infrastructure costs. To achieve the goal of Energy Efficiency we need: A well-dimensioned Data Center which uses the resources/components only when needed . The peak workloads are covered by outsorcing the computational needs. Project Overview Our point of view It is not a green approach ... ... A green approach is ...
  • 7. Virtualization technologies and Cloud computing play a key role in this context. Our proposal aims at: developing an integrated solution to monitor and control the energy consumption of all the Cloud Nodes. taking advantage of all the available information: energy consumption ratings of all the components, current wokload in each Cloud Node, data rates, computational resources location, etc. dynamically adapting the active hardware to the needs in terms of computing and communication resources which are required by the running services. The Green Cloud Project Overview Our point of view
  • 8. In four steps: obtaining information about the Cloud Nodes: components, workloads, computing resources locations, etc. generating policies to improve the energy consumption, using all the available information. applying this policies in the Cloud Nodes. determining whether the components of the Cloud fit the needs for the future workload or an increase / reduction of hardware is advisable. Green Scope: Generic issues not aiming at reducing the energy waste which are common to Cloud Computing infrastructures are not the focus of this proposal Project Overview Our point of view
  • 9. Our solution covers the green area . Nevertheless, a cloud is a cloud (i.e. for Amazon its cloud is a private one) so this approach is applicable to both sides (with some special features). The Company Staffs machines are part of the problem and part of the solution too. Project Overview Public or Private? Both Private Cloud (DataCenter) Public Cloud, i.e. company staffs machines
  • 10. Resources Layer ... Project Overview Architecture. Our point of view ESS = Energy Saving System NCD = Network Card Driver HM = Hardware Manager PM = Processes Manager PA = Power Analizer (optional) NIAC HM NIAC NIAC Measurements DataBase PM HM PM HM PM PA Service Service Service Service Service Service Layer Control Layer Execution DataBase Third Parties like Amazon EC2 Deployment Manager Learning Manager Learnt Policies NCD NCD NCD SNMP Agent SNMP Agent SNMP Agent ESS ESS ESS Processor Layer Processor Allocation Manager Cloud Controller Learnt Policies V.M. V.M. V.M. V.M. V.M.
  • 11. Oscar Trenado Mu単oz Innovation Manager Tlf: +34 983 54 99 11 +34 606 10 73 50 email: [email_address] Miguel ngel Pe単alvo Bedia Innovation Consultant Tlf: +34 983 54 99 11 e-mail: [email_address] Contact us!
  • 12. Energy Saving Softwarebased onCloud Computing for Adjustable Processing Environments ( ESSCCAPE ) The Green Cloud