-
公开(公告)号:GB2499545A
公开(公告)日:2013-08-21
申请号:GB201309307
申请日:2011-10-28
Applicant: IBM
Inventor: TRAEGER AVISHAY , YEHUDA SHMUEL BEN , YASSOUR BEN-AMI , FACTOR MICHAEL , BOROVIK ERAN , MACHULSKY ZORIK , METH KALMAN
IPC: G06F9/455
Abstract: Systems and methods for optimizing operation of a physical device in a virtualized computing environment are provided. The method comprises monitoring instructions issued by an application running on a virtual machine's (VM) operating system, wherein the VM is hosted by a hypervisor providing access to a physical device connected to a virtualized computing environment; wherein a device driver supported by the operating system issues one or more instructions to an emulated device supported by the hypervisor to control the physical device according to the issued instructions, wherein the emulated device's registers are implemented in memory of the hypervisor and the instructions issued by the device driver are stored in said registers, wherein the hypervisor handles the instructions, in response to learning that the instructions are stored in said registers.
-
公开(公告)号:AT470187T
公开(公告)日:2010-06-15
申请号:AT04813127
申请日:2004-12-06
Applicant: IBM
Inventor: MAKHERVAKS VADIM , MACHULSKY ZORIK , BIRAN GIORA , SHALEV LEAH
Abstract: An RNIC implementation that performs direct data placement to memory where all segments of a particular connection are aligned, or moves data through reassembly buffers where all segments of a particular connection are non-aligned. The type of connection that cuts-through without accessing the reassembly buffers is referred to as a "Fast" connection because it is highly likely to be aligned, while the other type is referred to as a "Slow" connection. When a consumer establishes a connection, it specifies a connection type. The connection type can change from Fast to Slow and back. The invention reduces memory bandwidth, latency, error recovery using TCP retransmit and provides for a "graceful recovery" from an empty receive queue. The implementation also may conduct CRC validation for a majority of inbound DDP segments in the Fast connection before sending a TCP acknowledgement (Ack) confirming segment reception.
-
公开(公告)号:CA2548966C
公开(公告)日:2010-06-01
申请号:CA2548966
申请日:2004-12-06
Applicant: IBM
Inventor: MAKHER-VAKS VADIM , MACHULSKY ZORIK , BIRAN GIORA , SHALEV LEAH
Abstract: An RNIC implementation that performs direct data placement to memory where all segments of a particular connection are aligned, or moves data through reassembly buffers where all segments of a particular connection are non-aligned. The type of connection that cuts-through without accessing the reassembly buffers is referred to as a "Fast" connection because it is highly likely to be aligned, while the other type is referred to as a "Slow" connection. When a consumer establishes a connection, it specifies a connection type (S2). The connection type can change from Fast to Slow and back. The invention reduces memory bandwidth, latency, error recovery using TCP retransmit and provides for a "graceful recovery" from an empty receive queue. The implementation also may conduct CRC validation (S11, S6) for a majority of inbound DDP segments in the Fast connection before sending a TCP acknowledgement (Ack) confirming segment reception.
-
公开(公告)号:DE112011103411T5
公开(公告)日:2013-08-01
申请号:DE112011103411
申请日:2011-10-28
Applicant: IBM
Inventor: TRAEGER AVISHAY , BEN YEHUDA SHMUEL , YASSOUR BEN-AMI , FACTOR MICHAEL , BOROVIK ERAN , MACHULSKY ZORIK , METH KALMAN
IPC: G06F9/455
Abstract: Es werden Systeme und Verfahren zum Optimieren des Betriebs einer physischen Einheit in einer virtualisierten Datenverarbeitungsumgebung bereitgestellt. Das Verfahren weist das Überwachen von Anweisungen auf, die durch eine auf dem Betriebssystem einer virtuellen Maschine (VM) ausgeführte Anwendung ausgegeben werden, wobei die VM durch einen Hypervisor betrieben wird, der den Zugriff auf eine physische Einheit bereitstellt, die mit einer virtualisierten Datenverarbeitungsumgebung verbunden ist; wobei ein durch das Betriebssystem unterstützter Einheitentreiber eine oder mehrere Anweisungen an eine durch den Hypervisor unterstützte emulierte Einheit ausgibt, um die physische Einheit gemäß den ausgegebenen Anweisungen zu steuern, wobei die Register der emulierten Einheit im Speicher des Hypervisor eingerichtet und die durch den Einheitentreiber ausgegebenen Anweisungen in den Registern gespeichert sind, wobei der Hypervisor die Anweisungen als Reaktion auf die Kenntnis abwickelt, dass die Anweisungen in den Registern gespeichert sind.
-
公开(公告)号:CA2548966A1
公开(公告)日:2005-07-07
申请号:CA2548966
申请日:2004-12-06
Applicant: IBM
Inventor: MAKHER-VAKS VADIM , MACHULSKY ZORIK , BIRAN GIORA , SHALEV LEAH
Abstract: An RNIC implementation that performs direct data placement to memory where a ll segments of a particular connection are aligned, or moves data through reassembly buffers where all segments of a particular connection are non- aligned. The type of connection that cuts-through without accessing the reassembly buffers is referred to as a "Fast" connection because it is highl y likely to be aligned, while the other type is referred to as a "Slow" connection. When a consumer establishes a connection, it specifies a connection type (S2). The connection type can change from Fast to Slow and back. The invention reduces memory bandwidth, latency, error recovery using TCP retransmit and provides for a "graceful recovery" from an empty receive queue. The implementation also may conduct CRC validation (S11, S6) for a majority of inbound DDP segments in the Fast connection before sending a TCP acknowledgement (Ack) confirming segment reception.
-
公开(公告)号:DE112011103411B4
公开(公告)日:2022-08-11
申请号:DE112011103411
申请日:2011-10-28
Applicant: IBM
Inventor: TRAEGER AVISHAY , BEN YEHUDA SHMUEL , YASSOUR BEN-AMI , FACTOR MICHAEL , BOROVIK ERAN , MACHULSKY ZORIK , METH KALMAN
IPC: G06F9/455
Abstract: Es werden Systeme und Verfahren zum Optimieren des Betriebs einer physischen Einheit in einer virtualisierten Datenverarbeitungsumgebung bereitgestellt. Das Verfahren weist das Überwachen von Anweisungen auf, die durch eine auf dem Betriebssystem einer virtuellen Maschine (VM) ausgeführte Anwendung ausgegeben werden, wobei die VM durch einen Hypervisor betrieben wird, der den Zugriff auf eine physische Einheit bereitstellt, die mit einer virtualisierten Datenverarbeitungsumgebung verbunden ist; wobei ein durch das Betriebssystem unterstützter Einheitentreiber eine oder mehrere Anweisungen an eine durch den Hypervisor unterstützte emulierte Einheit ausgibt, um die physische Einheit gemäß den ausgegebenen Anweisungen zu steuern, wobei die Register der emulierten Einheit im Speicher des Hypervisor eingerichtet und die durch den Einheitentreiber ausgegebenen Anweisungen in den Registern gespeichert sind, wobei der Hypervisor die Anweisungen als Reaktion auf die Kenntnis abwickelt, dass die Anweisungen in den Registern gespeichert sind.
-
公开(公告)号:CA2548085C
公开(公告)日:2014-04-29
申请号:CA2548085
申请日:2004-12-08
Applicant: IBM
Inventor: BIRAN GIORA , MACHULSKY ZORIK , MAKHERVAKS VADIM , SHALEV LEAH
Abstract: A network interface controller implementation that performs direct data placement to memory where all segments of a particular connection are aligned, i.e. cuts-through without accessing reassembly buffers (i.e., a "Fast" connection type), or if all segments are non~aligned moves data through reassembly buffers (i.e., a "Slow" connection type). The type of connection can change from Fast to Slow and back. The implementation preferably conducts CRC validation for a majority of inbound DDP segments in the Fast connection before sending a TCP acknowledgement confirming segment reception. The method of computing a CRC value according to the present invention assumes that each TCP segment starts with an aligned DDP segment and that the first two bytes of a TCP payload is an MPA length field of an MPA frame.
-
公开(公告)号:CA2548085A1
公开(公告)日:2005-07-07
申请号:CA2548085
申请日:2004-12-08
Applicant: IBM
Inventor: SHALEV LEAH , MAKHERVAKS VADIM , MACHULSKY ZORIK , BIRAN GIORA
Abstract: A network interface controller implementation that performs direct data placement to memory where all segments of a particular connection are aligne d, i.e. cuts-through without accessing reassembly buffers (i.e., a "Fast" connection type), or if all segments are non~aligned moves data through reassembly buffers (i.e., a "Slow" connection type). The type of connection can change from Fast to Slow and back. The implementation preferably conduct s CRC validation for a majority of inbound DDP segments in the Fast connection before sending a TCP acknowledgement confirming segment reception. The metho d of computing a CRC value according to the present invention assumes that eac h TCP segment starts with an aligned DDP segment and that the first two bytes of a TCP payload is an MPA length field of an MPA frame.
-
-
-
-
-
-
-