Systems, methods, and media for controlling delivery of content
Methods, systems, and computer readable media for controlling delivery of content are provided. In some embodiments, a system for controlling delivery of content is provided. The system includes processing circuitry configured to: transmit, to a server, a plurality of requests for blocks of the content; while at least some of the plurality of requests are still outstanding: detect a change of a service characteristic of a connection between the system and the server; determine a preferred number of outstanding requests; and cancel at least some of the requests from the plurality that are still outstanding based on the preferred number and a count of the requests from the plurality that are still outstanding.
1. An operating method for controlling delivery of content through a server, the operating method comprising:
receiving, from a device, a plurality of requests for blocks of the content;
wherein an actual number of outstanding requests is at least one;
obtaining a set of measurements of service characteristics, wherein each of the set of measurements of service characteristics comprises a measurement that is indicative of at least one of:
a latency of a connection between the server and the device, wherein the latency includes a time differential between a transmission of a request for a block of content from the device to the server and receipt of a first network packet associated with the block of content at the device; and
a bandwidth of a connection between the server and the device, wherein the bandwidth includes a size of the block divided by a time differential between the receipt of the first network packet and receipt of a last network packet associated with the block of content;
determining a preferred number of outstanding requests based upon the set of measurements of service characteristics and the size of the block of content;
transmitting, to the device, the preferred number of outstanding requests;
when a service characteristic of the set of measurements of service characteristics suggests at least one of an increase of the bandwidth and a decrease of the latency:
increasing the preferred number of outstanding requests; and
transmitting, to the device, the preferred number of outstanding requests; and
when an actual number of outstanding requests is less than the preferred number of outstanding requests, reducing the actual number of outstanding requests.
2. The operating method of claim 1 , wherein determining the preferred number of outstanding requests is evaluated using the following expression:
preferred
number
of
requests
=
⌈
t
latency
t
transmission
⌉
+
1
;
wherein t latency is the latency and t transmission is a transmission time.
3. The operating method of claim 2 , wherein the transmission time is evaluated using the following expression:
t
transmission
=
block
size
bandwidth
;
wherein block size is the size of the block of content currently being transmitted and bandwidth is the bandwidth.
4. The operating method of claim 1 , wherein the plurality of requests for blocks of content are transmitted using an Open Systems Interconnect (OSI) application-layer protocol.
5. The operating method of claim 1 , wherein each block of content requested in the plurality of requests for blocks of the content is a fragment of a larger content file.
6. The operating method of claim 1 , wherein the content is at least one of streaming content and content from a live performance.
7. The operating method of claim 1 , wherein reducing the actual number of outstanding requests comprises cancelling the outstanding requests in order of most recent.
8. The operating method of claim 1 , further comprising:
determining whether an underflow condition has occurred; and
when an underflow condition has occurred, reducing the actual number of outstanding requests in response to the underflow condition.
9. The operating method of claim 8 , wherein an underflow condition occurs when:
size
data
bandwidth
≥
t
video
in
buffer
-
t
delay
-
t
cancel
request
-
t
safety
margin
,
wherein:
size data is a combined size of all requests of the outstanding requests;
bandwidth is the bandwidth of the connection between the server and the device;
t video in buffer is a combined play time of all blocks of the content that are stored in a buffer of the device;
t delay is a minimum playtime of content data that needs to be stored in the buffer of the device in order to prevent an underflow;
t cancel request is an estimate of time needed for the device to cancel a request; and
T safety margin is an additional safety margin.
10. The operating method of claim 1 , wherein a service characteristic is selected from the group consisting of network throughput, a signal strength of a network connection of the device, and type of network access.
11. A server directed to controlling delivery of content, comprising at least one processor which is configured to:
receive, from a device, a plurality of requests for blocks of the content;
wherein an actual number of outstanding requests is at least one:
obtain a set of measurements of service characteristics,
wherein the set of measurements of service characteristics comprises one or more measurements that are indicative of at least one of:
a latency of a connection between the server and the device, wherein the latency includes a time differential between a transmission of a request for a block of content from the device to the server and receipt of a first network packet associated with the block of content at the device; and
a bandwidth of a connection between the server and the device, wherein the bandwidth includes a size of the block divided by a time differential between the receipt of the first network packet and receipt of a last network packet associated with the block of content;
determine a preferred number of outstanding requests based upon the set of measurements of service characteristics;
when a service characteristic of the set of measurements of service characteristics suggests at least one of an increase of the bandwidth and a decrease of the latency:
increase the preferred number of outstanding requests; and
transmit, to the device, the preferred number of outstanding requests; and
when an actual number of outstanding requests is less than the preferred number of outstanding requests, reduce the actual number of outstanding requests.
12. The server of claim 11 , wherein determining the preferred number of outstanding requests is evaluated using the following expression:
preferred
number
of
requests
=
⌈
t
latency
t
transmission
⌉
+
1
;
wherein t latency is the latency and t transmission , is a transmission time.
13. The server of claim 12 , wherein the transmission time is evaluated using the following expression:
t
transmission
=
block
size
bandwidth
;
wherein block size is the size of the block of content currently being transmitted and bandwidth is the bandwidth.
14. The server of claim 11 , wherein the plurality of requests for blocks of content are transmitted using an Open Systems Interconnect (OSI) application-layer protocol.
15. The server of claim 11 , wherein each block of content requested in the plurality of requests for blocks of the content is a fragment of a larger content file.
16. The server of claim 11 , wherein the content is at least one of streaming content and content from a live performance.
17. The server of claim 11 , wherein reducing the actual number of outstanding requests comprises cancelling the outstanding requests in order of most recent.
18. The server of claim 11 , wherein the at least one processor is further configured to:
determine whether an underflow condition has occurred; and
when an underflow condition has occurred, reduce the actual number of outstanding requests in response to the underflow condition.
19. The server of claim 18 , wherein an underflow condition occurs when:
size
data
bandwidth
≥
t
video
in
buffer
-
t
delay
-
t
cancel
request
-
t
safety
margin
,
wherein:
size data is a combined size of all requests of the outstanding requests;
bandwidth is the bandwidth of the connection between the server and the device;
t video in buffer is a combined play time of all blocks of the content that are stored in a buffer of the device;
t delay is a minimum playtime of content data that needs to be stored in the buffer of device in order to prevent an underflow;
t cancel request is an estimate of time needed for the device to cancel a request; and
T safety margin is an additional safety margin.
20. The server of claim 11 , wherein a service characteristic is selected from the group consisting of network throughput, a signal strength of a network connection of the device, and type of network access.