Choosing the Right Disk Technology in a High Availability Environment DRAFT Version 2.0, August 1996

HASS Chassis # 1 HASS Chassis # 2
Primary PrimaryMirror Mirror
DRAFT -- Revision 2.0
August 22, 1996Page 12
Figure 2: Typical High Availability Configuration with the High Availability Storage
System (HASS)
LVM Mirror Consistency
LVM has two mechanisms for maintaining consistency between the mirrors. The
default mechanism is to use the Mirror Write Cache (MWC). The MWC is kept in
memory and is periodically written to the disk as a Mirror Consistency Record (MCR) in
case of OS failure. Upon reboot, LVM knows which mirror copy is current and which
one is stale and will quickly resynchronize the mirror copies. Unfortunately, posting the
MCR to disk might degrade the performance of write-intensive applications. The exact
amount of degradation is extremely application dependent.
Disabling the MWC enables the alternate method of keeping the mirror copies
synchronized and is called NOMWC. Although enabling NOMWC might improve on-
line performance slightly, it greatly increases recovery time since the entire Logical
Volume must be copied from one side of the mirror to the other. This method might
also result in lost transactions since without the MWC, LVM has no reliable way of
choosing which side of the mirror is more current than the other side(s).
Mirrored JBODs and SCSI targets
A major disadvantage of JBODs is that each disk consumes a separate SCSI target
address. There is a hardware limit to the number of targets on a SCSI bus as shown in
the section on Disk Link Technologies. The F/W SCSI link provides the greatest
connectivity today, with a maximum of 15 targets. High availability clusters and
performance considerations reduce this to a maximum of 10 disk targets. Using 4 GB