Amazon Technologies, Inc.

Bradley D Roberts

Andrew Bruce Dickinson

Colin John Whittaker

Eric Paul Wei

Albert Park Niemoller

David John Ward, Jr

Frederick David Sinn

Edwin S Hong

Colin J Whittaker

Andrew B Dickinson

Albert P Niemoller

Eric P Wei

Systems and methods manage network traffic of a first protocol by use of a second protocol. Network traffic directed toward a network destination address of a first protocol is received. A mapping of addresses is utilized to determine a corresponding other network address of the second protocol. The network traffic is reconfigured to be forwarded to the intended network destination using the second protocol and the determined corresponding other network address.

Techniques for using mappings to manage network traffic

Network addresses are dynamically allocated to devices based on demand. It is determined that a quantity of virtualized machine instances executed on a plurality of computing devices has changed. An updated network address allocation is dynamically assigned to a subdivision of a network in response to the quantity of virtualized machine instances. The subdivision includes the plurality of computing devices. A router for the subdivision is reconfigured to implement the updated network address allocation.

Dynamically allocating network addresses

A stateless load balancer may route a request to a computing device without tracking state information of the communication connection that delivered the request. The stateless load balancer may receive a request that is initiated by a client device. The request may be transported to the stateless load balancer via a communication connection that includes a public destination internet protocol (IP) address. One of the routing devices in the stateless load balancer may obtain connection tuples of the communication connection that transported the request, in which the connection tuples include the public destination IP address. Once obtained, the routing device may generate a hash value from one or more connection tuples. Subsequently, the routing device may map the hash value to a target network address of a particular computing device in a cluster of multiple computing devices by accessing stored mapping data, and then route the request to the particular computing device for processing.

One-to-many stateless load balancing

Network addresses are dynamically allocated to devices based on demand. A first network address allocation is initially assigned to a subdivision of a network including multiple computing devices. A router for the subdivision is configured to implement the first network address allocation. Data indicating a reconfiguration of the computing devices is obtained. A second network address allocation is subsequently assigned to the subdivision of the network based at least in part on the reconfiguration of the computing devices. The router for the subdivision is reconfigured to implement the second network address allocation.

Instances executing within a programmable execution service ('PES') that are engaged in undesirable computing activity can be identified by comparing activity performed by instances executing within the PES to data describing known undesirable computing activity. Once compromised instances have been identified, other previously unknown undesirable computing activity performed by the compromised instances can be identified by determining whether activity is performed by the compromised instances significantly more often than other instances executing within the PES. New undesirable computing activity discovered using this process could then be utilized to discover other compromised instances.

Detecting undesirable computing activity

One-to many stateless load balancing

Methods and apparatus for capacity management of draining-state platforms providing network-accessible resources are disclosed. A system includes computing platforms configurable to host a plurality of resource instances, and a resource manager. The resource manager determines that a particular computing platform is in a draining state, wherein, in accordance with its designation as being in the draining state, the particular platform is currently hosting at least one uninterruptible resource instance and is temporarily unavailable to launch additional uninterruptible resource instances. In response to (a) a determination that the particular computing platform in the draining state has sufficient available computing capacity for at least one resource instance and (b) an instance activation request from a client, the resource manager launches an interruptible resource instance on the particular computing platform.

Growing community of inventors

Bellevue, WA, United States of America

Bradley D Roberts