INTER ZONE WRITE FOR ZONED NAMESPACES
An embodiment of an electronic apparatus may include one or more substrates, and logic coupled to the one or more substrates, the logic to control access to a persistent storage device which is logically organized with a namespace divided into two or more zones, receive a write request for a first zone of the two or more zones of the namespace, determine if the first zone can accommodate the write request, and span the write request across a second zone of the two or more zones of the namespace if the first zone cannot accommodate the write request. Other embodiments are disclosed and claimed.
1 . An electronic apparatus, comprising:
one or more substrates; and
logic coupled to the one or more substrates, the logic to:
control access to a persistent storage device which is logically organized with a namespace divided into two or more zones,
receive a write request for a first zone of the two or more zones of the namespace,
determine if the first zone can accommodate the write request, and
span the write request across a second zone of the two or more zones of the namespace if the first zone cannot accommodate the write request.
2 . The apparatus of claim 1 , wherein if the first zone cannot accommodate the write request the logic is further to determine if a zone adjacent to the first zone is uninitialized, and if so determined, the logic is further to:
initialize the adjacent zone to provide the second zone; and
continue the write request with the second zone.
3 . The apparatus of claim 2 , wherein if the adjacent zone is determined to be already initialized the logic is further to query a host to re-initialize the adjacent zone, and if the re-initialize query is accepted by the host, the logic is further to:
initialize the adjacent zone to provide the second zone; and
continue the write request with the second zone.
4 . The apparatus of claim 1 , wherein the logic is further to determine if the first zone is full, and if so determined, the logic is further to:
identify an empty zone;
reset the empty zone to provide the second zone; and
continue the write request with the second zone.
5 . The apparatus of claim 4 , wherein the logic is further to:
manage a pointer to link the second zone to the first zone.
6 . The apparatus of claim 4 , wherein the logic is further to:
query a host to identify the empty zone.
7 . The apparatus of claim 1 , wherein the persistent storage device comprises a solid state drive.
8 . An electronic storage system, comprising:
a persistent storage device logically organized with a namespace divided into two or more zones; and
a controller coupled to the persistent storage device, the controller including logic to:
receive a write request for a first zone of the two or more zones of the namespace,
determine if the first zone can accommodate the write request, and
span the write request across a second zone of the two or more zones of the namespace if the first zone cannot accommodate the write request.
9 . The system of claim 8 , wherein if the first zone cannot accommodate the write request the logic is further to determine if a zone adjacent to the first zone is uninitialized, and if so determined, the logic is further to:
initialize the adjacent zone to provide the second zone; and
continue the write request with the second zone.
10 . The system of claim 9 , wherein if the adjacent zone is determined to be already initialized the logic is further to query a host to re-initialize the adjacent zone, and if the re-initialize query is accepted by the host, the logic is further to:
initialize the adjacent zone to provide the second zone; and
continue the write request with the second zone.
11 . The system of claim 8 , wherein the logic is further to determine if the first zone is full, and if so determined, the logic is further to:
identify an empty zone;
reset the empty zone to provide the second zone; and
continue the write request with the second zone.
12 . The system of claim 11 , wherein the logic is further to:
manage a pointer to link the second zone to the first zone.
13 . The system of claim 11 , wherein the logic is further to:
query a host to identify the empty zone.
14 . The system of claim 8 , wherein the persistent storage device comprises a solid state drive.
15 . A method of controlling storage, comprising:
controlling access to a persistent storage device which is logically organized with a namespace divided into two or more zones;
receiving a write request for a first zone of the two or more zones of the namespace;
determining if the first zone can accommodate the write request; and
spanning the write request across a second zone of the two or more zones of the namespace if the first zone cannot accommodate the write request.
16 . The method of claim 15 , wherein if the first zone cannot accommodate the write request the method further comprises determining if a zone adjacent to the first zone is uninitialized, and if so determined, the method further comprising:
initializing the adjacent zone to provide the second zone; and
continuing the write request with the second zone.
17 . The method of claim 16 , wherein if the adjacent zone is determined to be already initialized the method further comprises querying a host to re-initialize the adjacent zone, and if the re-initialize query is accepted by the host, the method further comprising:
initializing the adjacent zone to provide the second zone; and
continuing the write request with the second zone.
18 . The method of claim 15 , further comprising determining if the first zone is full, and if so determined, the method further comprising:
identifying an empty zone;
resetting the empty zone to provide the second zone; and
continuing the write request with the second zone.
19 . The method of claim 18 , further comprising:
managing a pointer to link the second zone to the first zone.
20 . The method of claim 18 , further comprising:
querying a host to identify the empty zone.