Composite public cloud, method and system

Abstract

A composite public cloud enables an automated marketplace system for, selling and trading of computational resources by aggregating available compute resources so that buyers are automatically matched with sellers. A listing of compute resources available on a prospective seller's system and bids by prospective buyers for resources offered is maintained. A seller uses a series of attributes to characterize an available resource for prospective buyers. Compute resources are presented in a form which can be easily quantified or compartmentalized. Prospective buyers search the marketplace system based on desired resource criteria. The marketplace continually evaluates the attributes of resources available for sale and the desired criteria of prospective buyers. The system attempts to match buyers and sellers, optimizing for the needs of both parties. If a desirable match is found, the marketplace system facilitates a transaction between both parties, making the purchased resource available to the buyer.

1 . A system for creating a composite public cloud in a cloud-computing environment comprising:
 a bus;   a processor coupled to the bus; and   a memory medium coupled to the bus, the memory medium comprising instructions for:
 creating a plurality of tenants in at least one availability zone of at least one private cloud, wherein an availability zone comprises a physical instance of said private cloud and wherein a tenant comprises a virtual representation of certain compute resources situated within an availability zone; 
 allocating a portion of said tenants as a ‘resale’ tenant; 
 associating a description to each resale tenant; and 
 aggregating all of said resale tenants into a Composite Public Cloud. 
   
     
     
         2 . The system of  claim 1 , wherein said physical instance comprises a plurality of servers and associated storage, each of said servers having resident thereon an instance of a hypervisor application, said hypervisor application operative to virtualize said server so that each server is represented as at least one virtual machine; and
 wherein a tenant comprises at least one virtual machine.   
     
     
         3 . The system of  claim 1 , wherein aggregating all of said resale tenants into a Composite Public Cloud comprises:
 registering at least one resale tenant on a facility for transferring commoditized excess compute resources,   mapping each resale tenant to a public tenant within a public availability zone in a virtualized public cloud; and   publishing said description of said registered resale client.   
     
     
         4 . The system of  claim 1 , wherein associating a description to each resale tenant comprises:
 describing said resale tenant according to one or more predetermined attributes, said predetermined attributes including at least:   start of availability;   grace period;   cost; and   minimal duration.   
     
     
         5 . The system of  claim 1 , said memory further comprising instructions for:
 logically segregating computing resources from each other within the same cloud, to filter traffic between different members of the same cloud; and   specifying rules for how that traffic should flow.   
     
     
         6 . The system of  claim 1 , said memory further comprising instructions for:
 establishing a portal for buying and selling commoditized excess compute resources;   automatically matching buyers and sellers by means of said portal; and   conducting speculative trading of commoditized excess compute resources.   
     
     
         7 . A computer-implemented method for creating a composite public cloud in a cloud-computing environment comprising:
 a computing device creating a plurality of tenants in at least one availability zone of at least one private cloud, wherein an availability zone comprises a physical instance of said private cloud and wherein a tenant comprises a virtual representation of   certain compute resources situated within an availability zone;   a computing device allocating a portion of said tenants as ‘resale’ tenants;   a computing device associating a description to each resale tenant; and   a computing device aggregating at least some of said resale tenants into a Composite Public Cloud.   
     
     
         8 . The method of  claim 7 , wherein said physical instance comprises a plurality of servers and associated storage, each of said servers having resident thereon an instance of a hypervisor application, said hypervisor application operative to virtualize said server so that each server is represented as at least one virtual machine; and
 wherein a tenant comprises at least one virtual machine.   
     
     
         9 . The method of  claim 7 , wherein aggregating all of said resale tenants into a Composite Public Cloud comprises:
 a computing device registering at least one resale tenant on a portal for trading computing resources;   a computing device mapping each resale tenant to a public tenant within a public availability zone n said Composite Public Cloud; and   a computing device publishing said description of said registered resale client.   
     
     
         10 . The method of  claim 7 , wherein associating a description to each resale tenant comprises:
 a computing device describing said resale tenant according to one or more predetermined attributes, said predetermined attributes including at least:   start of availability;   grace period;   cost; and   minimal duration.   
     
     
         11 . The method of  claim 7 , further comprising:
 a computing device logically segregating computing resources from each other within the same cloud, to filter traffic between different members of the same cloud; and   a computing device specifying rules for how that traffic should flow.   
     
     
         12 . The method of  claim 7 , said memory further comprising instructions for:
 a computing device establishing a portal for buying and selling commoditized excess compute resources;   a computing device matching buyers and sellers by means of said portal; and   a computing device conducting speculative trading of commoditized excess compute resources.   
     
     
         13 . A system for brokering transfer of commoditized compute resources comprising:
 a bus;   a processor coupled to the bus; and   a memory medium coupled to the bus, the memory medium comprising instructions for:
 obtaining and maintaining an inventory of commoditized compute resources; 
 aggregating said inventory into a Composite Public Cloud; and 
 facilitating transfer of said commoditized compute resources between buyers and sellers. 
   
     
     
         14 . The system of  claim 13 , wherein obtaining and maintaining an inventory of commoditized compute resources comprises:
 a system broker managing a private cloud owned by a prospective seller of compute resources;   said system broker imposing measurement and quality constraints on compute resources within said private cloud to guarantee quality of resale compute resources;   continually monitoring performance data; and   metering compute consumption in the interest of performance control.   
     
     
         15 . The system of  claim 13 , wherein obtaining and maintaining an inventory of commoditized compute resources comprises;
 obtaining inventory of known quality from private clouds having excess compute resources;   mapping available inventory into sellable lots of compute over time, each sellable lot being described by at least one attribute selected from a group comprising:   start of availability;   grace period;   cost; and   minimal duration.   
     
     
         16 . The system of  claim 15 , wherein aggregating said inventory into a Composite Public Cloud comprises:
 mapping at least a portion of said commoditized computer resources to at least one public tenant of a public availability zone within said Composite Public Cloud; and   publishing a description of said commoditized computer resources.   
     
     
         17 . The system of  claim 15 , said memory further comprising instructions for:
 segregating and encapsulating newly sold inventory so as not to present security risk to private cloud tenants or prior sold inventory, wherein compute containment occurs by means of packet inspection and filtering at all layers of the network stack.   
     
     
         18 . The system of  claim 15 , said memory further comprising instructions for:
 tracking all contracts from lot-sold-time to delivery and the percentage of time a contract is delivered without interruption—up to the grace period, and   making said percentage of time available to the compute lot purchaser as a measure of quality.   
     
     
         19 . The system of  claim 15 , said memory further comprising instructions for:
 presenting available inventory as a Composite Public Cloud via a consumer-facing portal;   consumers using the portal to both purchase contracts and associated purchased contracts with virtual machines;   caching virtual machines and placing them in the associated Private Cloud at a contract's “Start of Availability”; and   responsive to inadvertent termination of a virtual machine, attempting to buy a contract for equal or lesser cost and duration and placing the last cached copy of the terminated virtual machine in the next viable private cloud.   
     
     
         20 . A computer-implemented method for brokering transfer of commoditized compute resources comprising:
 a computing device obtaining and maintaining an inventory of commoditized compute resources;   a computing device aggregating said inventory into a Composite Public or Private Cloud; and   a computing device facilitating transfer of said commoditized compute resources between buyers and sellers.   
     
     
         21 . The method of  claim 20 , further comprising:
 identifying excess compute resources within said private cloud;   designating said excess compute resources as resale compute resources;   commoditizing said resale compute resources;   offering said commoditized resale compute resources for sale via said Composite Private Cloud; and   receiving revenue from at least one purchaser of at least a portion of said commoditized resale compute resources offered for sale.   
     
     
         22 . The method of  claim 20 , further comprising:
 managing said private cloud by a provider for a subscriber, said managing comprising:
 responsive to a subscriber request, said provider designing and installing at least one data center for said subscriber; 
 said provider virtualizing said data center to least one private cloud owned by said subscriber; 
 said provider implementing within said private cloud at least one availability zone; 
 said provider allocating compute resources to said at least one availability zone; 
 said provider further allocating said resources allocated to said at least one availability zone to a plurality of tenants within said at least one availability zone, wherein said plurality of tenants includes at least one private tenant and at least one resale tenant; 
 segregating said at least one private tenant and said at least one resale tenant so as not to present security risk to private cloud tenants or purchasers of said at least one resale tenant; 
 said provider imposing measurement and quality constraints on compute resources within said private cloud to guarantee quality of resale compute resources; 
 said provider continually monitoring performance data; and 
 said provider metering compute consumption in the interest of performance control. 
   
     
     
         23 . The method of  claim 20 , further comprising:
 said computing device aggregating said inventory into at least one availability zone.   
     
     
         24 . The method of  claim 20 , said at least one availability zone comprising:
 a public cloud in which virtual machines are provided to buyers who pay for time-based compute resources, wherein different prices are set for any of CPU and memory.   
     
     
         25 . The method of  claim 20 , further comprising a plurality of availability zones, wherein different prices are set for at least two of said availability zones. 
     
     
         26 . The method of  claim 20 , said at least one availability zone comprising:
 a virtual private cloud, wherein hypervisor, network, and storage resources are grouped into an availability zone comprising a single private commoditized compute resource for a buyer.   
     
     
         27 . The method of  claim 20 , said computing device comprising:
 a cloud hosting service for aggregating said inventory into at least one availability zone in which resources are packaged as a pre-configured virtual private server, wherein said commoditized compute resources are grouped into packages that are sold on a plan billing basis.   
     
     
         28 . The method of  claim 20 , said computing device comprising:
 a service for aggregating said inventory into hybrid servers for providing said commoditized compute resources to said buyers by allocating hypervisors on a unitary basis, wherein each buyer is assigned a dedicated hypervisor for said buyer's service.   
     
     
         29 . The method of  claim 28 , wherein said inventory is provided as a backup to said allocated hypervisor. 
     
     
         30 . The method of  claim 20 , said computing device comprising:
 a cloud hosting service for aggregating said inventory to provide said buyers with a virtual private server using local storage.