A control server receives probe data from a plurality of data centers indicating measured latencies with a first IP address associated with an origin server. The control server sums the measured latencies of a first data center having a lowest measured latency and a second data center. When the sum is below a threshold value, the control server determines the IP address to be an anycast IP address and selects a proper subset of the plurality of data centers as proxying data centers for other data centers in the plurality of data centers. When the sum is not below the threshold value, the control server determines the IP address to not be an anycast IP address and selects the first data center having the lowest measure latencies as the proxying data center for other data centers in the plurality of data centers.

Examples described herein relate to selecting a compute node to execute the function based on latency of retrievals of function data blocks, wherein the function data blocks comprise at least a portion of an image of the function and the function data blocks are stored among two or more tiers of storage. In some examples, a latency of retrievals of function data blocks is based on storage of the function data blocks in devices of one or more storage tiers and retrieval latency to a memory utilized to launch the function.

The disclosed computer-implemented method includes determining that incoming media item requests are to be skewed from a random distribution among server nodes, using a random distribution algorithm, to a directed distribution among the server nodes. The method then includes identifying, in a loading assignment, which media items are to be loaded onto specific server nodes to produce the directed distribution of media item requests. The method next includes preloading the identified media items onto the server nodes according to the loading assignment and receiving media item requests for the preloaded media items. The method then includes routing the received media item requests to the server nodes using the random distribution algorithm, where the random distribution algorithm is skewed to the directed distribution based on the preloading of the media items according to the identified loading assignment. Various other methods, systems, and computer-readable media are also disclosed.

A fault handling method for a gateway server, including: in response to a first long connection request received from a client, selecting a first master node from a master load balancing cluster, wherein the first master node establishes a long connection between the client and a service server on the basis of the first long connection request and generates session information; and in response to a determination that a fault occurs in the first master node, performing the following operations: selecting a second master node from the master load balancing cluster; and selecting a first slave node from a slave load balancing cluster and controlling the second master node to send the data from the first master node to the first slave node such that the first slave node maintains the long connection between the client and the service server on the basis of the session information.

A fault handling method for a gateway server, including: in response to a first long connection request received from a client, selecting a first master node from a master load balancing cluster, wherein the first master node establishes a long connection between the client and a service server on the basis of the first long connection request and generates session information; and in response to a determination that a fault occurs in the first master node, performing the following operations: selecting a second master node from the master load balancing cluster; and selecting a first slave node from a slave load balancing cluster and controlling the second master node to send the data from the first master node to the first slave node such that the first slave node maintains the long connection between the client and the service server on the basis of the session information.

A method and apparatus of a device that downloads content to a local device using a decentralized content distribution network is described. In an exemplary embodiment, the device receives a request from an application to download content using the decentralized content distribution network. In addition, the device determines a plurality of nodes in the decentralized content distribution network to download the content. For each of the plurality of nodes, the device attempts to establish communications between the device and that node. In addition, the device downloads a portion of the content from that node if communication is established between the local device and that node, wherein the size of the portion is based on at least a characteristic of that node. Furthermore, the device downloads a portion of the content from a proxy if communication cannot be established between the local device and that node, wherein the size of the portion is based on at least a characteristic of that proxy.

Embodiments of the present disclosure relate to methods and apparatus for managing caching in mobile edge computing systems. According to an embodiment of the present disclosure, a method includes: determining a caching policy for caching a group of content items in a plurality of edge nodes, wherein determining the cache policy comprises determining whether to cache a content item in an edge node at least based on a popularity of the content item and a latency between the edge node and a group of base stations; and transmitting a content indication to each edge node of the plurality of edge nodes based on the determined caching polity respectively, herein the content indication transmitted to a respective edge node comprises one or more identifiers corresponding to one or more content items determined to be cached in the respective edge node.

Embodiments of the present disclosure relate to methods and apparatus for managing caching in mobile edge computing systems. According to an embodiment of the present disclosure, a method includes: determining a caching policy for caching a group of content items in a plurality of edge nodes, wherein determining the cache policy comprises determining whether to cache a content item in an edge node at least based on a popularity of the content item and a latency between the edge node and a group of base stations; and transmitting a content indication to each edge node of the plurality of edge nodes based on the determined caching polity respectively, herein the content indication transmitted to a respective edge node comprises one or more identifiers corresponding to one or more content items determined to be cached in the respective edge node.

The present invention relates to a method for enabling status updates in a 5G core network. The method is performed in a network function (NF), service producer node and comprises sending (S310) a hypermedia message to an NF repository function (NRF), wherein the hypermedia message comprises an NF status indication and a hypermedia link for query of current NF status of the service producer, receiving (S350) a hypermedia message from a service consumer by the hypermedia link, which hypermedia message request a current NF status of the service producer, and sending (S360) the current NF status to the service consumer in response to the received request.

Embodiments for scaling services in a distributed network by registering bucket names for the services with a Domain Name Service (DNS) server. Buckets hold resources and distribution of buckets to servers is maintained using a Consistent Hashing Ring. Upon determining that services servers need to be added or deleted based on changes load conditions, leave/join events are managed by a Kubernetes event service and updates are propagated to all of the services by the membership service component. A service scaling process updates the mapping of buckets to the host name of the new service server. A resource request from a client is sent to the mapped bucket name, and a DNS resolver process sends it to the changed service server so that the client is completely unaware of this change in bucket ownership.