Scalable, real-time messaging system
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for receiving from a plurality of publisher clients a plurality of messages, assigning each of the messages to one of a plurality of distinct channels wherein each channel comprises an ordered plurality of messages, storing messages of each of the channels in one or more respective buffers according to the order, each buffer having a respective time-to-live, retrieving messages for one or more of the channels from respective buffers having time-to-lives that have not expired and according to the order, and sending the retrieved messages to a plurality of subscriber clients.
1. A computer-implemented method, comprising:
receiving a plurality of messages from a plurality of publisher clients, each message associated with a particular one of a plurality of channels; and
storing each of the received messages in a respective buffer for the channel associated with the message and according to an order, the buffer having an expiration time based on when the buffer was allocated to the channel,
wherein retrieval of messages for the particular channel is from one or more buffers allocated to the channel that have not expired and according to the order.
2. The method of claim 1 , wherein storing each of the received messages comprises:
identifying a first buffer for storing messages of the particular channel;
ordering messages for the particular channel based on, at least in part, a respective time each of the messages was received from respective publisher clients; and
storing a first portion of the messages of the particular channel as ordered in the first buffer.
3. The method of claim 2 , further comprising:
determining that the first buffer is full and, based thereon:
identifying a second buffer; and
storing a different second portion of the messages of the particular channel as ordered in the second buffer.
4. The method of claim 2 , wherein storing the first portion of the messages further comprises:
buffering one or more of the first portion of the messages in a local data buffer; and
storing the buffered messages in the first buffer after a number of buffered messages exceeds a predetermined size or after a predetermined time period has elapsed.
5. The method of claim 1 , wherein the particular channel is initiated by a message from a publisher client.
6. The method of claim 1 , comprising:
retrieving messages for the particular channel by initiating a connection from a subscriber client subscribing to the particular channel.
7. The method of claim 6 , wherein retrieving messages comprises:
retrieving messages for the particular channel and storing the retrieved messages in a local buffer, wherein each message is removed from the local buffer when a respective time of publication of the message has exceeded a predetermined time period.
8. The method of claim 6 , comprising:
sending messages retrieved for the particular channel to one or more subscriber clients that have subscribed to the particular channel.
9. The method of claim 1 , wherein storing each of the received messages in the respective buffer for the associated channel and according to the order comprises:
storing messages of each channel earlier in the order in buffers having time-to-lives that will expire sooner than buffers used to store messages of each channel later in the order.
10. The method of claim 1 , wherein the order comprises the order in which messages for the channel were received.
11. A system comprising:
one or more computers programmed to perform operations comprising:
receiving a plurality of messages from a plurality of publisher clients, each message associated with a particular one of a plurality of channels; and
storing each of the received messages in a respective buffer for the channel associated with the message and according to an order, the buffer having an expiration time based on when the buffer was allocated to the channel,
wherein retrieval of messages for the particular channel is from one or more buffers allocated to the channel that have not expired and according to the order.
12. The system of claim 11 , wherein storing each of the received messages comprises:
identifying a first buffer for storing messages of the particular channel;
ordering messages for the particular channel based on, at least in part, a respective time each of the messages was received from respective publisher clients; and
storing a first portion of the messages of the particular channel as ordered in the first buffer.
13. The system of claim 12 , wherein the operations further comprise:
determining that the first buffer is full and, based thereon:
identifying a second buffer; and
storing a different second portion of the messages of the particular channel as ordered in the second buffer.
14. The system of claim 12 , wherein storing the first portion of the messages further comprises:
buffering one or more of the first portion of the messages in a local data buffer; and
storing the buffered messages in the first buffer after a number of buffered messages exceeds a predetermined size or after a predetermined time period has elapsed.
15. The system of claim 11 , wherein the particular channel is initiated by a message from a publisher client.
16. The system of claim 11 , wherein the operations further comprise:
retrieving messages for the particular channel by initiating a connection from a subscriber client subscribing to the particular channel.
17. The system of claim 16 , wherein retrieving messages comprises:
retrieving messages for the particular channel and storing the retrieved messages in a local buffer, wherein each message is removed from the local buffer when a respective time of publication of the message has exceeded a predetermined time period.
18. The system of claim 16 , wherein the operations further comprise:
sending messages retrieved for the particular channel to one or more subscriber clients that have subscribed to the particular channel.
19. The system of claim 11 , wherein storing each of the received messages in the respective buffer for the associated channel and according to the order comprises:
storing messages of each channel earlier in the order in buffers having time-to-lives that will expire sooner than buffers used to store messages of each channel later in the order.
20. An article comprising:
a non-transitory machine-readable medium having instructions stored thereon that when executed by one or more computers causes the computers to perform operations comprising:
receiving a plurality of messages from a plurality of publisher clients, each message associated with a particular one of a plurality of channels; and
storing each of the received messages in a respective buffer for the channel associated with the message and according to an order, the buffer having an expiration time based on when the buffer was allocated to the channel,
wherein retrieval of messages for the particular channel is from one or more buffers allocated to the channel that have not expired and according to the order.