Qdio Optimized Latency Mode; Large Send For Ipv6 Packets - IBM z13s Technical Manual

7.3.62 QDIO optimized latency mode

QDIO optimized latency mode (OLM) can help improve performance for applications that
have a critical requirement to minimize response times for inbound and outbound data.
OLM optimizes the interrupt processing in the following manner:
For inbound processing, the TCP/IP stack looks more frequently for available data to
process. This process ensures that any new data is read from the OSA-Express5S or
OSA-Express4S without needing more program controlled interrupts (PCIs).
For outbound processing, the OSA-Express5S or OSA-Express4S also look more
frequently for available data to process from the TCP/IP stack. Therefore, the process
does not require a Signal Adapter (SIGA) instruction to determine whether more data is
available.

7.3.63 Large send for IPv6 packets

Large send for IPv6 packets
segmentation processing from the host to an OSA-Express5S and OSA-Express4S feature
by employing a more efficient memory transfer into OSA-Express5S and OSA-Express4S.
Large send support for IPv6 packets applies to the OSA-Express5S and OSA-Express4S
features (CHPID type OSD and OSX), and is exclusive to z13, z13s, zEC12, zBC12, z196,
and z114 servers. With z13s, large send for IPv6 packets (segmentation offloading) for
LPAR-to-LPAR traffic is supported. The minimum support requirements are listed in
Table 7-50.
Table 7-50 Minimum support requirements for large send for IPv6 packets
Operating system
z/OS
z/VM
a. PTFs are required.
improves performance by offloading outbound TCP
Support requirements
a
z/OS V1R13
z/VM V6R2 for guest use only
Chapter 7. Software support
279