Adaptive transfer rate for retrieving content from a server

Abstract

An edge system receives requests from user devices to retrieves files from an origin server. Instead of retrieving the files as fast as possible, the edge system throttles the retrieval of files to a rate that just exceeds the speed at which the file is played by a browser or media player. The edge system determines an appropriate retrieval rate based on the contents of the file itself. For example, a manifest file associated with the file can indicate a time it takes to play back content and a bitrate of the content. Thus, the edge server can use this information to retrieve a file from an origin server at a rate that is just fast enough to minimize playback interruption. The retrieval rate determined by the edge server therefore does not rely on how fast or slow the user device retrieves the file from the edge server.

Description

BACKGROUND

A user may request a web page or other content page via a browser or media player operating on the user's computing device in order to stream content. For example, the browser may request content from a server, such as a streaming server. The streaming server may retrieve a file from an origin server that includes content that can be played by the browser, such as audio or video, and transmit the file to the browser. Before the file has been fully transmitted, the browser may begin playing the content included in the file. Once the content has been played, the file may be discarded by the device operating the browser.

If there is any delay in the transmission of the file such that the speed at which the file is played exceeds the speed at which the file is received, the user may be notified of the delay and playback may be paused. Various factors can contribute to this delay. These factors include, for example, (1) the speed of the wireless or wired connection between the user's device and the Internet, (2) the location of, and load on, the streaming server that provides the content, (3) the size of the requested content, and (4) the processing power of the user's device. When the delay is significant (e.g., several seconds or more), the task of playing content can be frustrating for the user.

BRIEF DESCRIPTION OF THE DRAWINGS

Throughout the drawings, reference numbers may be re-used to indicate correspondence between referenced elements. The drawings are provided to illustrate example embodiments described herein and are not intended to limit the scope of the disclosure.

FIG. 1 is a block diagram of a content retrieval environment that includes user devices, an edge system, and an origin server, according to one embodiment.

FIG. 2 is a block diagram of the content retrieval environment of FIG. 1 illustrating the operations performed by the components of the content retrieval environment to determine the fragment retrieval rate and retrieve the fragment, according to one embodiment.

FIG. 3 is a block diagram of the content retrieval environment of FIG. 1 illustrating the operations performed by the components of the content retrieval environment to determine the fragment retrieval rate, retrieve the fragment, and re-encode the fragment to avoid oscillating requests, according to one embodiment.

FIG. 4 is a block diagram of the content retrieval environment of FIG. 1 illustrating the operations performed by the components of the content retrieval environment to determine the rate for retrieving content from a user device and retrieve the content, according to one embodiment.

FIG. 5 is a flow diagram depicting a retrieval rate determination routine illustratively implemented by an edge system, according to one embodiment.

DETAILED DESCRIPTION

As described above, if there is any delay in the transmission of a file to a user device such that the speed at which the file is played exceeds the speed at which the file is received, playback may be paused, frustrating the user. Thus, when a user device requests a file from the streaming server, conventional streaming servers immediately request the file from the origin server at the fastest possible rate. While this minimizes situations in which the speed at which the file is played exceeds the speed at which the file is received, this also results in a wasteful usage of bandwidth by the streaming server. For example, some users start, but do not finish, consuming the requested content. Thus, conventional streaming servers waste network resources retrieving a complete file when only a portion of the file was needed. As another example, some users do not consume content in a sequential manner, but rather skip ahead or back to different portions of the content. Conventional streaming servers, by immediately requesting the file from the origin server, may end up retrieving a portion of the file that was not actually needed.

Wasting network bandwidth can prove costly for streaming server operators. For example, some streaming servers request content for thousands to hundreds of thousands of users at any given time. If usage at any given time is high or nearing bandwidth limits, retrieving unnecessary files can cause the retrieval of other, necessary files to be delayed, possibly resulting in the pausing of playback. As another example, some streaming server operators pay network operators for bandwidth usage according to the 95/5 model. In the 95/5 model, network operators may track the amount of bits transmitted or received in 5 minute periods (e.g., resulting in a bitrate corresponding to each 5 minute period). The bitrates falling in the top 5th percentile of all bitrates are dropped and the streaming server operator is charged based on the bitrate of the 95th percentile of all bitrates. If the streaming server is retrieving unnecessary files, this can increase the bitrate of the 95th percentile of all bitrates and therefore the amount charged by the network operator.

Generally described, various embodiments disclosed herein provide a streaming server (referred to herein as an edge system) that throttles the retrieval of files from the origin server to a rate that just exceeds the speed at which the file is played by a browser or media player. Instead of retrieving a file as fast as possible given network conditions, the edge system determines an appropriate retrieval rate based on the contents of the file itself. For example, content (e.g., audio, video, or audiovisual content) has an intrinsic time component. The time it takes to play back content, or a portion of the content, is known. A file size is also known. Thus, the edge server can use this information to retrieve a file from an origin server at a rate that is just fast enough to minimize playback interruption (e.g., 1.1 times faster than normal playback speed, 1.2 times faster than normal playback speed, etc.). The retrieval rate determined by the edge server therefore does not rely on how fast or slow the user device retrieves the file from the edge server. Additional details of the operations performed by the edge server in determining the retrieval rate are described in greater detail below with respect to FIGS. 1-5.

Example Content Retrieval Environment

FIG. 1 is a block diagram of a content retrieval environment that includes user devices 102, an edge system 104, and an origin server 106, according to one embodiment. As illustrated in FIG. 1, the content retrieval environment includes various user devices 102, various edge systems 104, and various origin servers 106. As will be appreciated by those of skill in the relevant art, the content retrieval environment may include any number of distinct user devices 102, edge systems 104, or origin servers 106. In other embodiments not shown, the content retrieval environment may also include other content sources, such as a content delivery network (“CDN”) server. The user devices 102 and the edge systems 104 may communicate with each other via one or more communication networks 110. The network 110 may be a publicly accessible network of linked networks, possibly operated by various distinct parties, such as the Internet. In other embodiments, the network 110 may include a private network, personal area network, local area network, wide area network, cable network, satellite network, cellular telephone network, etc. or combination thereof, each with access to and/or from the Internet. While not illustrated, the edge systems 104 and the origin servers 106 may also communicate via the network 110.

The edge system 104 includes a media fetcher 112, a client media receiver 114, a media re-encoder 116, and a media cache 118. In an embodiment, the media fetcher 112 retrieves files from the origin server 106 at the request of a user via a user device 102. The media fetcher 112 also determines a rate of retrieval from the origin server 106 based on the requested file. For example, the media fetcher 112 receives a request for a file from the user device 102. Specifically, the request indicates a fragment of the file to retrieve, which is typically the first fragment of the file if the user is beginning playback. In some cases, a file can be encoded at different bitrates and therefore the request may also include a desired bitrate. The media fetcher 112 queries the media cache 118 to determine whether the fragment at the desired bitrate is located therein. If the fragment at the desired bitrate is located in the media cache 118, then the media fetcher 112 retrieves the fragment from the media cache 118 and transmits the fragment to the user device 102 to complete the request. If the fragment at the desired bitrate is not located in the media cache 118, then the media fetcher 112 begins the retrieving the fragment from the origin server 106.

The media fetcher 112 can initially retrieve a manifest file associated with the requested file from the origin server 106. Alternatively, the manifest file may be stored in the media cache 118 and retrieved or may be retrieved from the user device 102. The manifest file indicates a playback duration of the content in the requested file or a playback duration of each fragment of the requested file. The manifest file also indicates a bitrate of the requested file or each fragment of the requested file. Using the playback duration information, the media fetcher 112 determines a maximum amount of time available to retrieve a fragment. To minimize playback interruptions, the media fetcher 112 determines a fragment retrieval time that is just faster than the maximum amount of time available to retrieve the fragment. For example, the fragment retrieval time may be 1.1× or 1.2× faster than the maximum amount of time available to retrieve the fragment.

The media fetcher 112 can use the fragment retrieval time, the playback duration information, and the bitrate information to determine a fragment retrieval bitrate. For example, the media fetcher 112 can multiply the playback duration of the fragment by the bitrate of the fragment to determine the bit size of the fragment. The media fetcher 112 can then divide the bit size of the fragment by the fragment retrieval time to determine the bitrate at which the fragment should be retrieved from the origin server 106. Thus, the fragment retrieval bitrate is not determined based on how long a user device 102 actually takes to retrieve and playback a fragment. The media fetcher 112 then retrieves the fragment from the origin server 106 at the fragment retrieval bitrate.

In other embodiments, the request for the fragment from the user device 102 also includes a user device 102 playback rate. For example, the user device 102 can play fragments at a rate faster or slower than the normal playback rate. This information can be provided to the media fetcher 112 as an additional factor in determining the fragment retrieval bitrate. For example, if the user device 102 is playing fragments at 1.5× speed, then the media fetcher 112 can use this information to determine an updated playback duration and use the updated playback duration to determine the maximum amount of time available to retrieve a fragment. The user device 102 playback rate can also indicate whether there are playback interruptions, such as a buffering event. The media fetcher 112 can estimate an updated playback duration based on this indication and use the updated playback duration to appropriately increase the maximum amount of time available to retrieve a fragment. The bit size determination may still be based on the normal or original playback duration. Alternatively, instead of receiving the user device 102 playback rate from the user device 102, the media fetcher 112 can determine the user device 102 playback rate based on a rate at which the user device 102 requests fragments.

In alternate embodiments, the media fetcher 112 uses information other than the manifest file to determine the playback duration information or the bitrate information. For example, the media fetcher 112 can determine this information by analyzing the raw file, based on a type of codec used to encode the content, by performing heuristics on the content (e.g., more solid colors may indicate a smaller file size and thus a lower bitrate), by analyzing metadata attached to a hypertext transfer protocol (HTTP) request, or by analyzing a request pattern of the user device 102 (e.g., the user device 102 generally requests fragments at a given bitrate), or the like.

Once a fragment is retrieved from the origin server 106, the media fetcher 112 stores the fragment in the media cache 118 and transmits the fragment to the user device 102. The media fetcher 112 can repeat the retrieval process described above for the next requested fragment.

Generally, the user device 102 begins playback of a fragment once the fragment is completely received. The user device 102 also sends a confirmation to the media fetcher 112 once a fragment has been completely received along with a request for the next fragment in sequence (or another fragment if the user skips ahead or behind). Thus, the media fetcher 112 sets the fragment retrieval bitrate such that the next fragment is retrieved from the origin server 106 before the user device 102 completes playback of the previous fragment. Accordingly, if fragments are associated with different playback times, then the media fetcher 112 uses the playback time associated with a previous fragment in determining the fragment retrieval bitrate for the next fragment.

In an embodiment, the retrieval process described above applies to each fragment after the initial or first couple of fragments of the requested file. For example, when playback begins and the user device 102 initially requests the file, the initial fragment or the first couple of fragments may be requested from the origin server 106 as fast as possible so that playback begins sooner (if such fragments are not already stored in the media cache 118). In addition, the user device 102 can notify the media fetcher 112 if a fragment buffer is nearly empty or empty and the media fetcher 112 can increase the rate at which subsequent fragments are retrieved from the origin server 106 (at least until the fragment buffer is no longer nearly empty or empty).

The media fetcher 112 can implement additional techniques to reduce bandwidth usage. For example, while the user device 102 request may specify a bitrate, the specified bitrate could be for content that is at a higher resolution than what is capable of being displayed on the user device 102. As an illustrative example, a mobile phone can request content at a bitrate corresponding to 4K video; however, the mobile phone may only be able to display content with resolutions up to 720p. Transmitting the fragment at the bitrate corresponding to 4K video as opposed to a fragment at a bitrate corresponding to 720p video is wasteful because the 4K fragment has a larger file size than the 720p fragment and the user device 102 is not capable of taking advantage of the higher resolution. Thus, the media fetcher 112 can request capabilities of the user device 102 and use this information to ignore the user device 102 request, if necessary, and select the appropriate fragment to retrieve from the media cache 118 or the origin server 106. Alternatively, the media re-encoder 116 or a separate component between the edge system 104 and the origin server 106 can receive the user device 102 capabilities from the media fetcher 112 and downconvert requested fragments to meet the capabilities of the user device 102 (e.g., downconvert from a resolution of 4K to 720p, downconvert from a stereo audio to mono audio, etc.).

Generally, user devices 102 request fragments at bitrates that would not cause playback interruptions (e.g., bitrates less than the available bandwidth of a network connection). In some embodiments, the available bandwidth of the network connection between the user device 102 and the media fetcher 112 fluctuates. Thus, for the same file, the user device 102 can request fragments at different bitrates as the available bandwidth fluctuates. For example, if the initial bandwidth of the network connection is 500 mb/s, the user device 102 may request a first fragment at 400 mb/s. If the available bandwidth then drops to 350 mb/s, the user device 102 may then request a second fragment at 325 mb/s. By retrieving fragments according to the retrieval process described above, the media fetcher 112 can avoid situations in which an entire file at a first bitrate is retrieved, only to be discarded later because the user device 102 later requests fragments at a second bitrate based on changed network conditions.

However, the user device 102 can oscillate between requesting a fragment at a high bitrate and a fragment at a low bitrate. Oscillating requests can cause playback interruption because the user device 102 buffer of fragments may empty earlier than expected. Thus, the media fetcher 112 can identify when such oscillations are occurring (e.g., the user device 102 has requested fragments at two different bitrates a threshold number of times within a certain period of time) and retrieve fragments at both bitrates, regardless of which bitrate the user device 102 request specifies. The fragment at the unrequested bitrate, though, can be retrieved at a slower rate than the fragment at the requested bitrate (but at a rate that is faster than the time to playback the previous fragment). Alternatively, the media fetcher 112 can identify when such oscillations are occurring and instruct the media re-encoder 116 to re-encode the fragment to a new bitrate that is low enough to avoid oscillating requests, as described in greater detail below.

In some embodiments, there are large gaps in bitrates that are available for some fragments in the origin server 106. Thus, when available network bandwidth fluctuates, the user device 102 may request fragments of widely different quality. For example, fragments may be available at 256 mb/s and at 720 mb/s. If the available network bandwidth fluctuates below and above 720 mb/s, then the user may experience a significant drop in quality when the available bandwidth drops below 720 mb/s and the user device 102 then requests a fragment at 256 mb/s. Thus, the media re-encoder 116 or another component between the edge system 104 and the origin server 106 can re-encode fragments as they are requested by the user device 102 to a bitrate that can reliably be requested by the user device 102 given network conditions (e.g., 500 mb/s in this example). The user device 102 can specifically request the re-encoded bitrate or the media fetcher 112 can request the higher bitrate fragment and instruct the media re-encoder 116 to re-encode the fragment to a more reliable bitrate.

In further embodiments, the media fetcher 112 can instruct the media re-encoder 116 to re-encode fragments for some, but not all, user devices 102 requesting the particular fragment. For example, the network bandwidth available to provide a fragment to a plurality of user devices 102 is limited. If the available network bandwidth is limited to the extent that a high bitrate fragment cannot be sent to all of the requesting user devices 102 or to the extent that different fragments cannot all be sent to the respective requesting user devices 102, then the media fetcher 112 can instruct the media re-encoder 116 to re-encode the high bitrate fragment to a lower bitrate fragment. The media fetcher 112 can then send the high bitrate fragment to some user devices 102 and the lower bitrate fragment to other user devices 102 such that all user devices 102 receive a version of the respective requested fragment. The media fetcher 112 can determine which user devices 102 receive a higher bitrate fragment based on which user devices 102 more consistently request the higher bitrate (e.g., the user devices 102 that more consistently request the higher bitrate would be more likely to receive the higher bitrate fragment), user device 102 information, fragment quality information, or the like.

In some cases, available network bandwidth may be limited such that a requested fragment cannot be provided to the user device 102 in a manner that avoids a playback interruption. Thus, if the edge system 104 (e.g., the media fetcher 112) determines that a fragment cannot be retrieved and provided to the user device 102 to avoid playback interruption, the media fetcher 112 can instruct the user device 102 to slow playback of a fragment to a rate that would allow the media fetcher 112 to provide the next fragment to the user device 102 in a manner that avoids a playback interruption (e.g., slowing playback to a 0.95× speed). The media fetcher 112 can determine the slower playback rate based on characteristics of the network bandwidth (e.g., an available upload or download bitrate) and characteristics of the fragment to be retrieved (e.g., the playback duration, the bitrate, etc.).

While the disclosure above is described with respect to audio, video, or audiovisual content, this is not meant to be limiting. For example, the techniques described herein can also be used to retrieve pages of a literary work (e.g., book, magazine, newspaper, etc.) or any other content for which continuous or consistent delivery or presentation is desired. Illustratively, each page of a literary work can be associated with an estimated reading time. The estimated reading time can be specific to a user (e.g., based on analyzing reading patterns of the user) or based on statistics of a plurality of users that have read the particular page (e.g., 90% of users read the page in 2 minutes, which can be set as the estimated reading time). The media fetcher 112 can then retrieve each page from the origin server 106 at a rate that is faster than the estimated reading time.

Furthermore, the techniques used to determine a rate of retrieval from the origin server 106 can also be used to determine a rate of retrieval from a user device 102. The client media receiver 114 is configured to receive content from the user device 102 or other devices via the network 110. Such content can include images, virus definitions, backups, data (e.g., sensor data) from Internet of Things (IoT) devices, audio, video, audiovisual, or the like. The user device 102 can specify a deadline by which transmission of the content to the edge system 104 needs to be complete. The client media receiver 114 can use the size of the content along with the amount of time remaining before the deadline is reached to determine the retrieval bitrate. For example, the client media receiver 114 can divide the size of the content by the amount of time remaining to determine a minimum bitrate. The client media receiver 114 can then retrieve the content at the minimum bitrate or at a bitrate that is just higher than the minimum bitrate (e.g., 1.1× higher, 1.2× higher, etc.). As with the retrieval of content from the origin server 106, throttling the retrieval of content from the user device 102 can also decrease streaming server operator costs.

The user devices 102 can include a wide variety of computing devices, including personal computing devices, terminal computing devices, laptop computing devices, tablet computing devices, electronic reader devices, mobile devices (e.g., mobile phones, media players, handheld gaming devices, etc.), wearable devices with network access and program execution capabilities (e.g., “smart watches” or “smart eyewear”), wireless devices, set-top boxes, gaming consoles, entertainment systems, televisions with network access and program execution capabilities (e.g., “smart TVs”), and various other electronic devices and appliances. Individual user devices 102 may execute media player application 120 to communicate via the network 110 with other computing systems, such as the host system 104, in order to request and play content.

The origin servers 106 (or CDNs server, not shown) can host and provide network-accessible content (e.g., images, audio, video, audiovisual, etc.). The origin servers 106 and CDN servers can correspond to logical associations of one or more computing devices for hosting content and servicing requests for the hosted content over the network 110. For example, an origin server 106 or CDN server can include a web server component corresponding to one or more server computing devices for obtaining and processing requests for content from the edge system 104 or other devices or service providers. In some embodiments, one or more origin servers 106 may be associated with one or more CDN service providers (e.g., entities that manage multiple CDN servers), application service providers, etc.

The edge system 104 may be a single computing device, or it may include multiple distinct computing devices, such as computer servers, logically or physically grouped together to collectively operate as a server system. The components of the edge system 104 can each be implemented in application-specific hardware (e.g., a server computing device with one or more ASICs) such that no software is necessary, or as a combination of hardware and software. In addition, the modules and components of the edge system 104 can be combined on one server computing device or separated individually or into groups on several server computing devices. In some embodiments, the edge system 104 may include additional or fewer components than illustrated in FIG. 1.

Example Block Diagram for Determining the Fragment Retrieval Rate

FIG. 2 is a block diagram of the content retrieval environment of FIG. 1 illustrating the operations performed by the components of the content retrieval environment to determine the fragment retrieval rate and retrieve the fragment, according to one embodiment. As illustrated in FIG. 2, the user device 102 requests a second fragment of a media file (1). In an embodiment, the user device 102 requests the second fragment after receiving the first fragment of the media file from the media fetcher 112. After receiving the first fragment, the user device 102 plays the first fragment of the media file (2). Alternatively, the user device 102 can begin playing the first fragment before requesting the second fragment.

The media fetcher 112 receives the request for the second fragment and requests the second fragment from the media cache 118 (3). In some embodiments, the media cache 118 stores the second fragment and returns a copy of the second fragment to the media fetcher 112. However, as illustrated in FIG. 2, the media cache 118 indicates that the second fragment is not available (4). Thus, the media fetcher begins to retrieve the second fragment from the origin server 106.

The media fetcher 112 can request the manifest file of the media file (5) from the origin server 106. The media fetcher 112 requests the manifest file if the manifest file is not already stored on the edge system 104. For example, the manifest file may already be stored on the edge system 104 because the media fetcher 112 earlier retrieved the first fragment. The origin server 106 then transmits the manifest file (6) to the media fetcher 112.

The media fetcher 112 determines a second fragment retrieval rate based on the bitrate and the second fragment duration included in the manifest file (7). Alternatively, the second fragment retrieval rate can be based on the bitrate and the first fragment duration included in the manifest file (e.g., the duration of the fragment before the second fragment).

Once the retrieval rate is determined, the media fetcher 112 requests the second fragment to be transmitted at the determined retrieval rate (8). The origin server 106 then proceeds to transmit the second fragment at the determined retrieval rate (9). Once retrieved from the origin server 106, the media fetcher 112 forwards the second fragment (10) to the user device 102 to complete the initial request.

Example Block Diagram for Determining the Retrieval Rate and Re-Encoding a Fragment

FIG. 3 is a block diagram of the content retrieval environment of FIG. 1 illustrating the operations performed by the components of the content retrieval environment to determine the fragment retrieval rate, retrieve the fragment, and re-encode the fragment to avoid oscillating requests, according to one embodiment. As illustrated in FIG. 3, the user device 102 requests a new fragment of a media file at a second bitrate (1). In an embodiment, the user device 102 requests the new fragment after receiving a first fragment of the media file at a first bitrate from the media fetcher 112. The first bitrate and the second bitrate are different, where the second bitrate is higher than the first bitrate. After receiving the first fragment, the user device 102 plays the first fragment of the media file (2). Alternatively, the user device 102 can begin playing the first fragment before requesting the new fragment.

The media fetcher 112 determines that the user device 102 is consistently requesting fragments of a first bitrate or a second bitrate (3). For example, the user device 102 may request fragments of the first bitrate rate or the second bitrate consistently over a certain period of time. Thus, assuming the manifest file is already retrieved, the media fetcher 112 determines the retrieval rate based on the second bitrate and the new fragment duration included in the manifest file (4). The media fetcher 112 can use the second bitrate, rather than the first bitrate, to determine the retrieval rate because the second bitrate is higher.

Once the retrieval rate is determined, the media fetcher 112 requests the new fragment to be transmitted at the determined retrieval rate (5). The origin server 106 then proceeds to transmit the new fragment at the determined retrieval rate (6).

The media fetcher 112 then requests the media re-encoder 116 to re-encode the new fragment at a third bitrate between the first bitrate and the second bitrate (7). For example, the third bitrate can be selected such that the user device 102 no longer has to request a lower bitrate fragment when the available network bandwidth fluctuates (e.g., the bitrate can be just below the bottom range of the bandwidth fluctuation). The media re-encoder 116 then re-encodes the new fragment and transmits the re-encoded new fragment (8) to the media fetcher 112. Once the re-encoded new fragment is received, the media fetcher 112 forwards the re-encoded new fragment (9) to the user device 102 to complete the initial request.

Example Block Diagram for Retrieving Content from a User Device

FIG. 4 is a block diagram of the content retrieval environment of FIG. 1 illustrating the operations performed by the components of the content retrieval environment to determine the rate for retrieving content from a user device 102 and retrieve the content, according to one embodiment. As illustrated in FIG. 4, the user device 102 requests the transfer of a media file (1). The request can include a size of the media file and deadline by which the transfer must be complete. The media file can include images, literary works, video, audio, or audiovisual material. The user device 102 can also request the transfer of other files, such as backup files, sensor data, or the like.

The client media receiver 114 receives the request and determines the media file retrieval rate based on the size of the media file and the transfer deadline (2). For example, the client media receiver 114 can divide the size of the media file by the time remaining before the deadline is reached to determine a minimum retrieval bitrate. The media file retrieval rate can be set to the minimum retrieval bitrate or to a bitrate slightly faster than the minimum retrieval bitrate (1.1× or 1.2× faster).

The client media receiver 114 can then transmit a request to the user device 102 to transfer the media file at the determined media file retrieval rate (3). Once the instruction is received, the user device 102 transmits the media file at the determined media file retrieval rate (4). After the transfer is complete, the client media receiver 114 can store the media file (5) in a media storage database 418 in association with the user or the user device 102.

Example Retrieval Rate Determination Routine

FIG. 5 is a flow diagram depicting a retrieval rate determination routine 500 illustratively implemented by an edge system, according to one embodiment. As an example, the edge system 104 (e.g., the media fetcher 112) of FIG. 1 can be configured to execute the retrieval rate determination routine 500. The retrieval rate determination routine 500 begins at block 502.

At block 504, a request for a fragment of a media file is received from a user device. For example, the fragment can be a portion of the media file that corresponds with a playback duration.

At block 506, a determination is made whether the fragment is stored in cache. If the fragment is stored in the cache, the retrieval rate determination routine 500 proceeds to block 508. Otherwise, the retrieval rate determination routine 500 proceeds to block 510.

At block 508, the fragment is retrieved from the cache. The retrieval rate determination routine 500 then proceeds to block 516.

At block 510, a retrieval rate for the fragment is determined based on contents of a manifest file associated with the media file. For example, the manifest file can include a playback duration associated with the fragment and a bitrate of the fragment. The playback duration can be used to determine a maximum amount of time available to retrieve the fragment. As an example, the media fetcher 112 can aim to retrieve the file 1.1× or 1.2× faster than the playback duration. Thus, the retrieval time, along with the bitrate of the fragment and the playback duration, can be used to determine the retrieval rate.

At block 512, the fragment is requested from the origin server at a transfer rate that is at least the retrieval rate. For example, the retrieval rate can be the minimum retrieval rate for requesting the fragment to reduce the likelihood that the user device 102 finishes playback of a previous fragment before the requested fragment is retrieved. The fragment can then be received from the origin server at the requested transfer rate, as illustrated at block 514.

At block 516, a determination is made whether to re-encode the fragment. For example, the fragment can be re-encoded if the user device 102 is consistently oscillating between requesting a high bitrate fragment and a low bitrate fragment. The fragment can also be re-encoded if the available network bandwidth between the edge system 104 and various user devices 102 is limited to the extent that not all user devices 102 can receive fragments at requested bitrates. If the fragment is to be re-encoded, the retrieval rate determination routine 500 proceeds to block 518. Otherwise, the retrieval rate determination routine 500 proceeds to block 520.

At block 518, the fragment is re-encoded. The fragment can be re-encoded to a bitrate lower than a requested bitrate, but higher than a bitrate available at the origin server. The retrieval rate determination routine 500 then proceeds to block 520.

At block 520, the fragment or the re-encoded fragment is transmitted to the user device. Once the transmission is complete, the user device may request another fragment and the routine 500 can be repeated. After the fragment is transmitted to the user device, the retrieval rate determination routine 500 may be complete, as shown in block 522.

All of the methods and tasks described herein may be performed and fully automated by a computer system. The computer system may, in some cases, include multiple distinct computers or computing devices (e.g., physical servers, workstations, storage arrays, cloud computing resources, etc.) that communicate and interoperate over a network to perform the described functions. Each such computing device typically includes a processor (or multiple processors) that executes program instructions or modules stored in a memory or other non-transitory computer-readable storage medium or device (e.g., solid state storage devices, disk drives, etc.). The various functions disclosed herein may be embodied in such program instructions, and/or may be implemented in application-specific circuitry (e.g., ASICs or FPGAs) of the computer system. Where the computer system includes multiple computing devices, these devices may, but need not, be co-located. The results of the disclosed methods and tasks may be persistently stored by transforming physical storage devices, such as solid state memory chips and/or magnetic disks, into a different state. In some embodiments, the computer system may be a cloud-based computing system whose processing resources are shared by multiple distinct business entities or other users.

Depending on the embodiment, certain acts, events, or functions of any of the processes or algorithms described herein can be performed in a different sequence, can be added, merged, or left out altogether (e.g., not all described operations or events are necessary for the practice of the algorithm). Moreover, in certain embodiments, operations or events can be performed concurrently, e.g., through multi-threaded processing, interrupt processing, or multiple processors or processor cores or on other parallel architectures, rather than sequentially.

The various illustrative logical blocks, modules, routines, and algorithm steps described in connection with the embodiments disclosed herein can be implemented as electronic hardware (e.g., ASICs or FPGA devices), computer software that runs on computer hardware, or combinations of both. Moreover, the various illustrative logical blocks and modules described in connection with the embodiments disclosed herein can be implemented or performed by a machine, such as a processor device, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A processor device can be a microprocessor, but in the alternative, the processor device can be a controller, microcontroller, or state machine, combinations of the same, or the like. A processor device can include electrical circuitry configured to process computer-executable instructions. In another embodiment, a processor device includes an FPGA or other programmable device that performs logic operations without processing computer-executable instructions. A processor device can also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration. Although described herein primarily with respect to digital technology, a processor device may also include primarily analog components. For example, some or all of the rendering techniques described herein may be implemented in analog circuitry or mixed analog and digital circuitry. A computing environment can include any type of computer system, including, but not limited to, a computer system based on a microprocessor, a mainframe computer, a digital signal processor, a portable computing device, a device controller, or a computational engine within an appliance, to name a few.

The elements of a method, process, routine, or algorithm described in connection with the embodiments disclosed herein can be embodied directly in hardware, in a software module executed by a processor device, or in a combination of the two. A software module can reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of a non-transitory computer-readable storage medium. An exemplary storage medium can be coupled to the processor device such that the processor device can read information from, and write information to, the storage medium. In the alternative, the storage medium can be integral to the processor device. The processor device and the storage medium can reside in an ASIC. The ASIC can reside in a user terminal. In the alternative, the processor device and the storage medium can reside as discrete components in a user terminal.

Conditional language used herein, such as, among others, “can,” “could,” “might,” “may,” “e.g.,” and the like, unless specifically stated otherwise, or otherwise understood within the context as used, is generally intended to convey that certain embodiments include, while other embodiments do not include, certain features, elements and/or steps. Thus, such conditional language is not generally intended to imply that features, elements and/or steps are in any way required for one or more embodiments or that one or more embodiments necessarily include logic for deciding, with or without other input or prompting, whether these features, elements and/or steps are included or are to be performed in any particular embodiment. The terms “comprising,” “including,” “having,” and the like are synonymous and are used inclusively, in an open-ended fashion, and do not exclude additional elements, features, acts, operations, and so forth. Also, the term “or” is used in its inclusive sense (and not in its exclusive sense) so that when used, for example, to connect a list of elements, the term “or” means one, some, or all of the elements in the list.

Disjunctive language such as the phrase “at least one of X, Y, or Z,” unless specifically stated otherwise, is otherwise understood with the context as used in general to present that an item, term, etc., may be either X, Y, or Z, or any combination thereof (e.g., X, Y, and/or Z). Thus, such disjunctive language is not generally intended to, and should not, imply that certain embodiments require at least one of X, at least one of Y, and at least one of Z to each be present.

While the above detailed description has shown, described, and pointed out novel features as applied to various embodiments, it can be understood that various omissions, substitutions, and changes in the form and details of the devices or algorithms illustrated can be made without departing from the spirit of the disclosure. As can be recognized, certain embodiments described herein can be embodied within a form that does not provide all of the features and benefits set forth herein, as some features can be used or practiced separately from others. The scope of certain embodiments disclosed herein is indicated by the appended claims rather than by the foregoing description. All changes which come within the meaning and range of equivalency of the claims are to be embraced within their scope.

Claims

1. A system comprising: a media cache configured to store fragments of content files; anda server system comprising a processor and memory, wherein the memory includes instructions that, when executed by the processor, cause the server system to: receive, from a user device, a request for a first fragment of a first content file;determine whether the first fragment is stored in the media cache;retrieve a manifest file associated with the first content file from an origin server in response to a determination that the first fragment is not stored in the media cache, wherein the manifest file comprises a playback time of the first fragment and a bitrate of the first fragment;determine a retrieval rate for the first fragment, wherein the retrieval rate is determined based, at least in part, on the playback time of the first fragment and the bitrate of the first fragment;request the first fragment from the origin server at a transfer rate that is at least equal to the determined retrieval rate;receive the first fragment from the origin server; andtransmit the received first fragment to the user device.

2. The system of claim 1, wherein the instructions further cause the server system to: determine a retrieval time of the first fragment based on the playback time of the first fragment and a time multiplier;determine a file size of the first fragment based on the playback time of the first fragment and the bitrate of the first fragment; anddetermine the retrieval rate based on the retrieval time and the file size.

3. The system of claim 1, wherein the instructions further cause the server system to determine the retrieval rate regardless of a time taken by the user device to play back a fragment retrieved prior to the first fragment.

4. The system of claim 1, wherein the request for the first fragment comprises a desired bitrate of the first fragment.

5. The system of claim 1, wherein the instructions further cause the server system to retrieve the first fragment from the media cache in response to a determination that the first fragment is stored in the media cache.

6. A computer-implemented method comprising: as implemented by one or more computing devices configured with specific executable instructions, receiving, from a user device, a request for a first fragment of a first content file;determining whether the first fragment is stored in a media cache;determining a retrieval rate for the first fragment in response to a determination that the first fragment is not stored in the media cache, wherein the retrieval rate is determined based, at least in part, on a playback time of the first fragment and a bitrate of the first fragment;requesting the first fragment from a server at a transfer rate that is at least the determined retrieval rate;receiving the first fragment from the server; andtransmitting the received first fragment to the user device.

7. The computer-implemented method of claim 6, further comprising retrieving a manifest file associated with the first content file from the server in response to the determination that the first fragment is not stored in the media cache, wherein the manifest file comprises the playback time of the first fragment and the bitrate of the first fragment.

8. The computer-implemented method of claim 7, wherein determining a retrieval rate for the first fragment further comprises: determining a retrieval time of the first fragment based on the playback time of the first fragment and a time multiplier;determining a file size of the first fragment based on the playback time of the first fragment and the bitrate of the first fragment; anddetermining the retrieval rate based on the retrieval time and the file size.

9. The computer-implemented method of claim 6, wherein determining a retrieval rate further comprises determining the retrieval rate regardless of a time taken by the user device to play back a fragment retrieved prior to the first fragment.

10. The computer-implemented method of claim 6, wherein the request for the first fragment comprises a desired bitrate of the first fragment.

11. The computer-implemented method of claim 10, wherein determining whether the first fragment is stored in a media cache further comprises: determining whether the first fragment at the desired bitrate is stored in the media cache; andretrieving a manifest file associated with the first content file from the server in response to a determination that the first fragment at the desired bitrate is not stored in the media cache.

12. The computer-implemented method of claim 10, further comprising: re-encoding the first fragment to a third bitrate between the desired bitrate and the second bitrate; andtransmitting the re-encoded first fragment to the user device.

13. The computer-implemented method of claim 6, further comprising retrieving the first fragment from the media cache in response to a determination that the first fragment is stored in the media cache.

14. The computer-implemented method of claim 6, wherein the content file comprises one of an audio file, a video file, an audiovisual file, or a literary work file.

15. A non-transitory computer-readable medium having stored thereon executable program code that directs a media fetcher operating on one or more computing devices to perform operations comprising: receiving, from a user device, a request for a first fragment of a first content file;determining that the first fragment is not stored in a media cache;determining a retrieval rate for the first fragment in response to the determination that the first fragment is not stored in the media cache, wherein the retrieval rate is determined based, at least in part, on a playback time of the first fragment and a bitrate of the first fragment;requesting the first fragment from a server at a transfer rate that is at least the determined retrieval rate;receiving the first fragment from the server; andtransmitting the received first fragment to the user device.

16. The non-transitory computer-readable medium of claim 15, wherein the operations further comprise retrieving a manifest file associated with the first content file from the server, wherein the manifest file comprises the playback time of the first fragment and the bitrate of the first fragment.

17. The non-transitory computer-readable medium of claim 15, wherein the operations further comprise: determining a retrieval time of the first fragment based on the playback time of the first fragment and a time multiplier;determining a file size of the first fragment based on the playback time of the first fragment and the bitrate of the first fragment; anddetermining the retrieval rate based on the retrieval time and the file size.

18. The non-transitory computer-readable medium of claim 15, wherein the operations further comprise determining the retrieval rate regardless of a time taken by the user device to play back a fragment retrieved prior to the first fragment.

19. The non-transitory computer-readable medium of claim 15, wherein the operations further comprise: determining whether the first fragment at a desired bitrate is stored in the media cache; andretrieving a manifest file associated with the first content file from the server in response to a determination that the first fragment at the desired bitrate is not stored in the media cache.

20. The non-transitory computer-readable medium of claim 15, wherein the operations further comprise: determining that the user device oscillates between requesting fragments at a desired bitrate and fragments at a second bitrate less than the desired bitrate;re-encoding the first fragment to a third bitrate between the desired bitrate and the second bitrate; andtransmitting the re-encoded first fragment to the user device.

21. The non-transitory computer-readable medium of claim 15, wherein the first content file comprises one of an audio file, a video file, an audiovisual file, or a literary work file.