SPARC Enterprise Technology -Business continuity

Index

• SPARC64™ VI carry forward mainframe high reliability technology  
  SPARC64 VI processors incorporated in SPARC Enterprise mid-range and high-end models, carry forward RAS (Reliability, Availability and Serviceability) technologies developed and proven on enterprise-class mainframes.
• Reliability of UltraSPARC T1 Processor 
  UltraSPARC T1 processors have high-level RAS functions (Reliability, Availability, Serviceability). Data in level1 cache memory is protected by parity check and level2 cache memory and registers by ECC(Error Checking and Correction).
• Memory mirror and memory patrol for improved system reliability   
  SPARC Enterprise incorporates a memory redundancy mechanism. It improves single system reliability by protecting all data in memory, even from uncorrectable multi-bits errors.
• Main LSI interface data protection  
  Major LSI (System Controllers, Memory Access Controllers, Crossbars, I/O controllers, etc.) in SPARC Enterprise are Fujitsu manufactured components. The interfaces between these components, both data and address information, are protected by ECC.
• High reliability is embedded in all components  
  SPARC Enterprise systems continue to operate even in cases of component failure. Basic components (HDD, Power Supply Units, Fans etc) are redundant and hot-swap. CPUs can be dynamically downgraded.
• Server monitoring and administration by Extended System Control Facility (XSCF)  
  SPARC Enterprise M4000 models and above, are monitored and administered by the XSCF incorporated in each server cabinet. XSCF incorporates a service processor separate from all other processors.
• Server monitoring and administration of SPARC Enterprise T series 
  SPARC Enterprise T2000 and T1000 are monitored and administered by Sun Advanced Lights Out Manager (ALOM) incorporated in the server cabinet. ALOM incorporates a service processor separate from all other processors.
• Whole system reliability supported by PRIMECLUSTER   
  High reliability of the total system is assured by the self-governing mechanisms within PRIMECLUSTER. These include, virtualization of redundant components, fast error detection, application take-over and system downgrading to ensure operational continuity.