In many networking implementations a high-speed packet buffer is used to store packets during periods of congestion. Usually this buffer is very small because it is implemented on-chip. Being a scarce and expensive resource, it is important to maximize the use of such buffers. It would be very desirable to implement methods and devices that address at least some of the shortcomings of the prior art.
Some implementations provide a method that involves the following steps: receiving a first set of inputs comprising a first plurality of entities and a first traffic behavior; determining a first region of a buffer corresponding to the first traffic behavior; assigning the first plurality of entities to the first region; determining hierarchical relationships between at least some of the first plurality of entities; determining a first shared buffer space of the first region; and assigning at least one threshold for each of the first plurality of entities. The threshold may comprise a maximum amount of the first shared buffer space that may be allocated to an entity. The method may also involve configuring a logic device to allocate the first shared buffer space dynamically according to the hierarchical relationships and the thresholds.
Reference will now be made in detail to some specific examples of the invention, including the best modes contemplated by the inventor for carrying out the invention. Examples of these specific embodiments are illustrated in the accompanying drawings. While the invention is described in conjunction with these specific embodiments, it will be understood that it is not intended to limit the invention to the described embodiments. On the contrary, it is intended to cover alternatives, modifications, and equivalents as may be included within the spirit and scope of the invention as defined by the appended claims.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. Particular example embodiments of the present invention may be implemented without some or all of these specific details. In other instances, well-known process operations have not been described in detail in order not to obscure unnecessarily the present invention.
Various techniques and mechanisms of the present invention will sometimes be described in singular form for clarity. However, it should be noted that some embodiments include multiple iterations of a technique or multiple instantiations of a mechanism unless noted otherwise. For example, a system uses a processor in a variety of contexts. However, it will be appreciated that a system can use multiple processors while remaining within the scope of the present invention unless otherwise noted.
Similarly, the steps of the methods shown and described herein are not necessarily all performed (and in some implementations are not performed) in the order indicated. Moreover, some implementations of the methods discussed herein may include more or fewer steps than those shown or described.
Furthermore, the techniques and mechanisms of the present invention will sometimes describe a connection between two entities. It should be noted that a connection between two entities does not necessarily mean a direct, unimpeded connection, as a variety of other entities may reside between the two entities. For example, a processor may be connected to memory, but it will be appreciated that a variety of bridges and controllers may reside between the processor and memory. Consequently, an indicated connection does not necessarily mean a direct, unimpeded connection unless otherwise noted.
Some examples of establishing buffer allocation schemes will be described with reference to
An entity may be a physical entity, such as a port or a physical interface. An entity may also be a logical entity, such as a virtual lane, a virtual output queue (“VOQ”) or a priority such as class of service (“CoS”) or quality of service (“QoS”).
Traffic behaviors can be simple or more complex, depending on the desired implementation. Examples include no drop, no drop with Backward Congestion Notification (“BCN”), drop with tail drop parameters specified, drop with BCN and per class flow control, drop with BCN, drop with AFD (Approximate Fair Dropping), etc. Many terms used herein, including but not limited to terms involving traffic behaviors, pertain to Data Center Ethernet (“DCE”) functionality.
A brief overview of some DCE concepts is provided in the following paragraphs. More detail regarding DCE may be found in United States Patent Application Publication No. 2006/0251067, entitled “FIBRE CHANNEL OVER ETHERNET”, which is hereby incorporated by reference. Methods and devices for managing network traffic, particularly in the DCE context, are described in United States Patent Application Publication No. 2006/0171318, entitled “ACTIVE QUEUE MANAGEMENT METHODS AND DEVICES” and in United States Patent Application Publication No. 2007/0081454, entitled “METHODS AND DEVICES FOR BACKWARD CONGESTION NOTIFICATION,” which are hereby incorporated by reference.
Although DCE switches of the present invention can be implemented using fiber optics and optical transceivers, some preferred DCE switches are implemented using copper connectivity to reduce costs. Some such implementations are implemented according to the proposed IEEE 802.3ak standard called 10Base-CX4, which is hereby incorporated by reference for all purposes. The inventors expect that other implementations will use the emerging standard IEEE P802.3an (10GBASE-T), which is also incorporated by reference for all purposes.
Servers 210 are also connected with DCE switch 245, which includes FC gateway 270 for communication with disk arrays 250. FC gateway 270 implements FC over Ethernet, thereby eliminating the need for separate FC and Ethernet networks within data center 200. Gateway 270 could be a device such as Cisco Systems' MDS 9000 IP Storage Service Module that has been configured with software for performing some methods of the present invention. Ethernet traffic is carried within data center 200 as native format. This is possible because DCE is an extension to Ethernet that can carry FC over Ethernet and RDMA in addition to native Ethernet.
In order to implement both behaviors on the same physical link 315, DCE involves methods and devices for implementing “virtual lanes,” also referred to herein as “VLs.” VLs allow a physical link to be divided into multiple logical entities such that traffic in one of the VLs is unaffected by the traffic on other VLs. This may be done by maintaining separate buffers (or separate portions of a physical buffer) for each VL. For example, it is possible to use one VL to transmit control plane traffic and some other high priority traffic without being blocked because of low priority bulk traffic on another VL. VLANs may be grouped into different VLs such that traffic in one set of VLANs can proceed unimpeded by traffic on other VLANs.
In the example illustrated by
It is preferable to dynamically partition the link into VLs, because static partitioning is less flexible. In some preferred implementations of the invention, dynamic partitioning is accomplished on a packet-by-packet basis (or a frame-by-frame basis), e.g., by adding an extension header or using specific fields of the frame header. For example, some implementations are based on the IEEE 802.1Q format. The present invention encompasses a wide variety of formats for such a header. In some implementations of the invention, there are two types of frames sent on a DCE link: these types are data frames and link management frames.
In this implementation, buffering is performed on both the input and output sides. Other architectures are possible, e.g., those having input buffers, output buffers and shared memory. Accordingly, each of input line cards 405 includes at least one buffer 410 and each of output line cards 425 includes at least one buffer 430, which may be any convenient type of buffer known in the art, e.g., an external DRAM-based buffer or an on-chip SRAM-based buffer. The buffers 410 are used for input buffering, e.g., to temporarily retain packets while awaiting sufficient buffer to become available at the output line card to store the packets to be sent across switching fabric 450. Buffers 430 are used for output buffering, e.g., to temporarily retain packets received from one or more of the input line cards 405 while awaiting sufficient credits for the packets to be transmitted to another DCE switch.
It is worthwhile noting that while credits may be used internally to a switch and also externally, there is not necessarily a one-to-one mapping between internal and external credits. Moreover, it is possible to use PAUSE frame either internally or externally. For example, any of the four possible combinations PAUSE-PAUSE, PAUSE-CREDITS, CREDITS-PAUSE and CREDIT-CREDIT may produce viable solutions.
DCE switch 400 includes some form of credit mechanism for exerting flow control. This flow control mechanism can exert back pressure on buffers 410 when an output queue of one of buffers 430 has reached its maximum capacity. For example, prior to sending a frame, one of the input line cards 405 may request a credit from arbiter 440 (which may be, e.g., a separate chip located at a central location or a set of chips distributed across the output linecards) prior to sending a frame from input queue 415 to output queue 435. Preferably, the request indicates the size of the frame, e.g., according to the frame credit field of the DCE header. Arbiter 440 will determine whether output queue 435 can accept the frame (i.e., output buffer 430 has enough space to accommodate the frame). If so, the credit request will be granted and arbiter 440 will send a credit grant to input queue 415. However, if output queue 435 is too full, the request will be denied and no credits will be sent to input queue 415.
DCE switch 400 needs to be able to support both the “drop” and “no drop” behavior required for virtual lanes, as discussed elsewhere herein. The “no drop” functionality is enabled, in part, by applying internally to the DCE switch some type of credit mechanism like the one described above. Externally, the “no drop” functionality can be implemented in accordance with the buffer-to-buffer credit mechanism described earlier or PAUSE frames. For example, if one of input line cards 405 is experiencing back pressure from one or more output line cards 425 through the internal credit mechanism, the line card can propagate that back pressure externally in an upstream direction via a buffer-to-buffer credit system like that of FC.
Preferably, the same chip (e.g., the same ASIC) that is providing “no drop” functionality will also provide “drop” functionality like that of a classical Ethernet switch. Although these tasks could be apportioned between different chips, providing both drop and no drop functionality on the same chip allows DCE switches to be provided at a substantially lower price.
Each DCE packet will contain information, e.g., in the DCE header, indicating the virtual lane to which the DCE packet belongs. DCE switch 400 will handle each DCE packet according to whether the VL to which the DCE packet is assigned is a drop or a no drop VL.
In addition to the DCE context, the present invention may be implemented in a variety of other contexts, including but not limited to Fibre Channel (“FC”) and Ethernet. Broadly speaking, FC would be considered a “no drop” protocol and Ethernet would be considered a “drop” protocol. Even within the FC and Ethernet contexts, many permutations of traffic behavior may be used, e.g., for flow control, to implement fairness, etc.
Returning now to
Such apportionment can provide various advantages. For example, apportioning a buffer into separate “drop” and “no drop” buffer regions avoids the challenges involved with having, e.g., a drop portion of the buffer competing with a no-drop portion over a shared buffer space. Such apportionment allows compatible fairness rules to be established and thresholds to be set in a global manner within each buffer region. However, as will be explained in more detail below, it is not always the case that an entity is associated with only one type of traffic behavior. A port, for example, is an entity that may be used to implement only one type of traffic behavior (e.g., in a simple Ethernet context) or several types of traffic behavior (e.g., in a DCE context). Therefore, the same entity may sometimes be mapped to more than one buffer region. In the DCE context, for example, the same port may be mapped to a drop buffer region and a no drop buffer region.
Some implementations, as here, may implement a hierarchical buffer allocation scheme. If so, entity hierarchies will be established. (Step 125.) For example, referring to tree diagram 500 of
Buffer regions preferably include reserved portions, as established in step 130 in this example, in order to provide at least a minimum amount of buffer space for each entity. The amount of buffer space reserved may or may not be the same for each entity at a particular level of a hierarchy. If an entity has child nodes, the amount of buffer space reserved for an entity may be determined by the sum of the amounts of buffer space reserved for each child node.
It may be desirable to establish latency buffer space, which is a buffer space dedicated to flow control. For buffer regions used to support a flow control mechanism, it is often desirable to have a specific portion of the buffer reserved for packets in transit because it takes a certain amount of time for a source to receive and react to a flow control indication. In the meantime, there are packets coming in. No-drop regions, for example, should have a latency buffer space large enough to ensure that no packets are dropped.
Therefore, in step 135 it is determined whether it would be desirable to establish a latency buffer portion for a buffer region. This determination may involve several factors, including but not limited to a determination of whether a region will be a drop or a no drop region: in the latter case, latency buffer space should be established. However, as described below with reference to
If it is determined in step 135 that it would be desirable to establish latency buffer space for a buffer region, the space is established. In this example, a max_latency threshold, which is the maximum amount of latency buffer, is first established for each entity in the region that requires a latency buffer. (Step 140.) The entire latency buffer portion of the region may then be determined. (Step 145.)
Preferably, shared portions are established in buffer regions. (Step 150.) In this example, the shared portion of each buffer region is the portion remaining after subtracting the reserved portion and the latency portion, if any. Here, a max_shared threshold is established for each entity that will share buffer space with other entities at the same level. (Step 155.) Such behavior will sometimes be referred to herein as “shared behavior.” As discussed in further detail below, the max_shared threshold may be established as a constant or as a variable. If the max_shared threshold is established as a constant, this will be referred to herein as “static” shared behavior (or the like). If the max_shared threshold is established as a variable, this will be referred to herein as “adaptive” shared behavior (or the like). In either case, shared behavior involves dynamic processes, wherein the instantaneous amount of buffer that an entity is using may change over time.
In some implementations, an adaptive max_shared threshold can be updated, e.g., based on the activity level of entities sharing the buffer space. For example, an adaptive max_shared threshold may be updated according to, e.g., the number of active entities at a particular level of the hierarchy at a particular time. Alternatively, an adaptive max_shared threshold may be updated according to an average number of active entities at a particular level of the hierarchy during a period of time.
In some implementations, an adaptive max_shared threshold may be updated according to a weighted number of active entities. For example, if entity A and entity B are at the same level of a hierarchy, a weighting function may be used to cause entity A's max_shared threshold to be N times larger than the max_shared threshold of entity B, given the same amount of activity. This may be desirable, for example, if one wants to assign more buffer space to an entity having a relatively higher class of service (“CoS”) than to an entity having a relatively lower CoS.
In step 160, it is determined whether additional thresholds should be defined for at least some entities. If so, the thresholds are established in step 165. For example, thresholds may be defined for asserting and de-asserting flow control, for implementing BCN, etc. The process ends in step 170.
Further examples of establishing buffer allocation schemes will now be described with reference to
Tree diagram 600 of
Similar regions of a buffer 700 are indicated in
However, not all entities share buffer space 710 in the same manner. In
In drop region 725, virtual lanes VL4 through VL7 may share portions of shared buffer 710 that is allocated to ports P0 through P7. Moreover, ports P0 through P7 can share portions of shared buffer 710 with one another. For example, ports P0 through P7 may be adaptively assigned buffer space according to the relative activity of their associated virtual lanes.
Buffer 800 of
In order to implement the pause mechanism, within drop region 725 portions of latency buffer space 715 are reserved for virtual lanes 4 and 5. In turn, virtual lanes 4 and 5 can share these portions of latency buffer space 715. Similarly, virtual lanes 2 and 3 can share reserved portions of latency buffer space 715 within no drop region 720. However, virtual lanes 0 and 1 do not share: they have a dedicated portion of latency buffer space 715. Moreover, unlike the example depicted in
Flow chart 900 of
In step 905, first through Nth activity levels are determined for first through Nth entities having adaptive shared behavior. For example, the activity levels of ports P0 through P7 in shared buffer portion 710 of drop region 725 in
In some such examples, step 905 may involve determining first through Mth activity levels of first through Mth sub-entities. Here, for example, step 905 involves a determination of how many of the virtual lanes corresponding to each of ports P0 through P7 have been active during a predetermined time. In this example, it is determined that ports P1 and P7 have relatively higher activity levels (with 3 and 4 active virtual lanes, respectively) as compared to ports P0 and P2 through P6 (having either no virtual lanes or 1 virtual lane active).
In step 910, it is determined whether there is an indication to change one or more max_shared values, according to a predetermined rule set. For example, step 910 may involve a determination as to how long the activity levels in question have persisted. In some implementations, for example, max_shared values will not be changed until activity levels have persisted for another predetermined length of time.
In some implementations, max_shared values may only be adjusted within a certain range. Here, for example, the applicable rule set may indicate that the max_shared value for each of ports P0 through P7 is limited to no more than, e.g., 25% of the shared buffer portion 710 in drop region 725. Similarly, the applicable rule set may indicate that the max_shared value for each of ports P0 through P7 must be at least a minimum value, e.g., 5% of the shared buffer portion 710 in drop region 725. Accordingly the determination of step 910 may also involve an evaluation of the present max_shared values of the entities involved and whether these present values can be increased or decreased.
In this example, each of ports P0 through P7 currently has a max_shared value corresponding to 15% of the shared buffer portion 710 of drop region 725. Moreover, the relative activity levels have persisted for a predetermined length of time. Therefore, the max_shared values will be adjusted in step 915. In this example, the max_shared values of port P1 and P7 are adjusted to 20% and 25%, respectively, of the shared buffer portion 710 of drop region 725. In this example, the adjustment is determined by starting with the minimum max_shared value of 5%, then adding 5% for each active entity. The remaining 55% is apportioned between ports P0 and P2 through P6, according to their relative activity levels.
This process may continue as needed. It is determined in step 920 whether to continue. In some implementations, the max_shared values of sub-entities may also be adaptive. Accordingly, process 900 may involve an evaluation of whether to change the max_shared values of, e.g., the virtual lanes corresponding to ports PI and P7.
The foregoing methods, along with other methods of the present invention, may be implemented by software, firmware and/or hardware. For example, at least some methods of the present invention may be implemented by computer programs embodied in machine-readable media, the software including instructions for controlling one or more devices to perform the method(s). Some aspects of the invention can be implemented by network devices or portions thereof.
Although illustrative embodiments and applications of this invention are shown and described herein, many variations and modifications are possible which remain within the concept, scope, and spirit of the invention, and these variations would become clear to those of ordinary skill in the art after perusal of this application. For example, although a substantial part of the foregoing discussion pertains to DCE networks, the implementations of the invention may be used in many other networks, including but not limited to Fibre Channel networks and high-speed Ethernet. Accordingly, the present embodiments are to be considered as illustrative and not restrictive, and the invention is not to be limited to the details given herein, but may be modified within the scope and equivalents of the appended claims.
Number | Name | Date | Kind |
---|---|---|---|
5402416 | Cieslak et al. | Mar 1995 | A |
5526350 | Gittins et al. | Jun 1996 | A |
5742604 | Edsall et al. | Apr 1998 | A |
5920566 | Hendel et al. | Jul 1999 | A |
5946313 | Allan et al. | Aug 1999 | A |
5974467 | Haddock et al. | Oct 1999 | A |
5987011 | Toh | Nov 1999 | A |
6021124 | Haartsen | Feb 2000 | A |
6104699 | Holender et al. | Aug 2000 | A |
6195356 | Anello et al. | Feb 2001 | B1 |
6201789 | Witkowski et al. | Mar 2001 | B1 |
6333917 | Lyon et al. | Dec 2001 | B1 |
6363477 | Fletcher et al. | Mar 2002 | B1 |
6397260 | Wils et al. | May 2002 | B1 |
6404768 | Basak et al. | Jun 2002 | B1 |
6414939 | Yamato | Jul 2002 | B1 |
6456590 | Ren et al. | Sep 2002 | B1 |
6459698 | Achrya | Oct 2002 | B1 |
6504836 | Li et al. | Jan 2003 | B1 |
6529489 | Kikuchi et al. | Mar 2003 | B1 |
6556541 | Bare | Apr 2003 | B1 |
6556578 | Silberschatz et al. | Apr 2003 | B1 |
6560198 | Ott et al. | May 2003 | B1 |
6587436 | Vu et al. | Jul 2003 | B1 |
6636524 | Chen et al. | Oct 2003 | B1 |
6650623 | Varma et al. | Nov 2003 | B1 |
6671258 | Bonneau | Dec 2003 | B1 |
6721316 | Epps et al. | Apr 2004 | B1 |
6724725 | Dreyer et al. | Apr 2004 | B1 |
6839794 | Schober | Jan 2005 | B1 |
6839858 | James et al. | Jan 2005 | B1 |
6885633 | Mikkonen | Apr 2005 | B1 |
6888824 | Fang et al. | May 2005 | B1 |
6901593 | Aweya et al. | May 2005 | B2 |
6904507 | Gil | Jun 2005 | B2 |
6934256 | Jacobson et al. | Aug 2005 | B1 |
6934292 | Ammitzboell | Aug 2005 | B1 |
6975581 | Medina et al. | Dec 2005 | B1 |
6975593 | Collier et al. | Dec 2005 | B2 |
6990529 | Yang et al. | Jan 2006 | B2 |
6999462 | Acharaya | Feb 2006 | B1 |
7016971 | Recio et al. | Mar 2006 | B1 |
7020715 | Venkataraman et al. | Mar 2006 | B2 |
7046631 | Giroux et al. | May 2006 | B1 |
7046666 | Bollay et al. | May 2006 | B1 |
7047666 | Hahn et al. | May 2006 | B2 |
7093024 | Craddock et al. | Aug 2006 | B2 |
7133405 | Graham et al. | Nov 2006 | B2 |
7133416 | Chamdani | Nov 2006 | B1 |
7158480 | Firoiu et al. | Jan 2007 | B1 |
7187688 | Garmire et al. | Mar 2007 | B2 |
7190667 | Susnov et al. | Mar 2007 | B2 |
7197047 | Latif et al. | Mar 2007 | B2 |
7209478 | Rojas et al. | Apr 2007 | B2 |
7209489 | Bailey et al. | Apr 2007 | B1 |
7221656 | Aweya et al. | May 2007 | B1 |
7225364 | Carnevale et al. | May 2007 | B2 |
7266122 | Hogg et al. | Sep 2007 | B1 |
7266598 | Rolia | Sep 2007 | B2 |
7277391 | Aweya et al. | Oct 2007 | B1 |
7286485 | Ouelette et al. | Oct 2007 | B1 |
7319669 | Kunz et al. | Jan 2008 | B1 |
7349334 | Rider | Mar 2008 | B2 |
7349336 | Mathews et al. | Mar 2008 | B2 |
7359321 | Sindhu et al. | Apr 2008 | B1 |
7385997 | Gorti et al. | Jun 2008 | B2 |
7400590 | Rygh et al. | Jul 2008 | B1 |
7400634 | Higashitaniguchi et al. | Jul 2008 | B2 |
7406092 | Dropps et al. | Jul 2008 | B2 |
7436845 | Rygh et al. | Oct 2008 | B1 |
7486689 | Mott | Feb 2009 | B1 |
7525983 | Dropps et al. | Apr 2009 | B2 |
7529243 | Sodder et al. | May 2009 | B2 |
7561571 | Lovett et al. | Jul 2009 | B1 |
7564789 | Betker | Jul 2009 | B2 |
7564869 | Cafiero et al. | Jul 2009 | B2 |
7684326 | Nation et al. | Mar 2010 | B2 |
7801125 | Kreeger et al. | Sep 2010 | B2 |
7830793 | Gai et al. | Nov 2010 | B2 |
20010043564 | Bloch et al. | Nov 2001 | A1 |
20010048661 | Clear et al. | Dec 2001 | A1 |
20020046271 | Huang | Apr 2002 | A1 |
20020085493 | Pekkala et al. | Jul 2002 | A1 |
20020085565 | Ku et al. | Jul 2002 | A1 |
20020103631 | Feldmann et al. | Aug 2002 | A1 |
20020141427 | McAlpine | Oct 2002 | A1 |
20020150041 | Reinschmidt et al. | Oct 2002 | A1 |
20020159385 | Susnow et al. | Oct 2002 | A1 |
20020188648 | Aweya et al. | Dec 2002 | A1 |
20020191640 | Haymes et al. | Dec 2002 | A1 |
20030002517 | Takajitsuko et al. | Jan 2003 | A1 |
20030026267 | Obermann et al. | Feb 2003 | A1 |
20030037127 | Shah et al. | Feb 2003 | A1 |
20030037163 | Kitada et al. | Feb 2003 | A1 |
20030061379 | Craddock et al. | Mar 2003 | A1 |
20030084219 | Yao et al. | May 2003 | A1 |
20030091037 | Latif et al. | May 2003 | A1 |
20030115355 | Cometto et al. | Jun 2003 | A1 |
20030118030 | Fukuda | Jun 2003 | A1 |
20030152034 | Zhang et al. | Aug 2003 | A1 |
20030152063 | Giese et al. | Aug 2003 | A1 |
20030169690 | Mott | Sep 2003 | A1 |
20030193942 | Gil | Oct 2003 | A1 |
20030195983 | Krause | Oct 2003 | A1 |
20030202536 | Foster et al. | Oct 2003 | A1 |
20030223416 | Rojas et al. | Dec 2003 | A1 |
20030225549 | Shay | Dec 2003 | A1 |
20030227893 | Bajic | Dec 2003 | A1 |
20040006705 | Walker | Jan 2004 | A1 |
20040008675 | Basso et al. | Jan 2004 | A1 |
20040013088 | Gregg | Jan 2004 | A1 |
20040013124 | Peebles et al. | Jan 2004 | A1 |
20040024903 | Costatino et al. | Feb 2004 | A1 |
20040032856 | Sandstrom | Feb 2004 | A1 |
20040042448 | Lebizay et al. | Mar 2004 | A1 |
20040042477 | Bitar et al. | Mar 2004 | A1 |
20040076175 | Patenaude | Apr 2004 | A1 |
20040078621 | Talaugon et al. | Apr 2004 | A1 |
20040081203 | Sodder et al. | Apr 2004 | A1 |
20040100980 | Jacobs et al. | May 2004 | A1 |
20040120332 | Hendel | Jun 2004 | A1 |
20040156390 | Prasad et al. | Aug 2004 | A1 |
20040196809 | Dillinger et al. | Oct 2004 | A1 |
20040213243 | Lin et al. | Oct 2004 | A1 |
20040240459 | Lo et al. | Dec 2004 | A1 |
20050002329 | Luft et al. | Jan 2005 | A1 |
20050018606 | Dropps et al. | Jan 2005 | A1 |
20050025179 | McLaggan et al. | Feb 2005 | A1 |
20050047421 | Solomon | Mar 2005 | A1 |
20050060445 | Beukema et al. | Mar 2005 | A1 |
20050104217 | Lehave | May 2005 | A1 |
20050105474 | Metzler | May 2005 | A1 |
20050111487 | Matta et al. | May 2005 | A1 |
20050138243 | Tierney et al. | Jun 2005 | A1 |
20050141419 | Bergamasco et al. | Jun 2005 | A1 |
20050141568 | Kwak et al. | Jun 2005 | A1 |
20050152406 | Chauveau | Jul 2005 | A2 |
20050169188 | Cometto et al. | Aug 2005 | A1 |
20050169270 | Mutuo et al. | Aug 2005 | A1 |
20050190752 | Chiou et al. | Sep 2005 | A1 |
20050226149 | Jacobson et al. | Oct 2005 | A1 |
20050238064 | Winter et al. | Oct 2005 | A1 |
20050286436 | Flask | Dec 2005 | A1 |
20060002385 | Johnsen et al. | Jan 2006 | A1 |
20060023708 | Snively | Feb 2006 | A1 |
20060059213 | Evoy | Mar 2006 | A1 |
20060087989 | Gai et al. | Apr 2006 | A1 |
20060098589 | Kreeger et al. | May 2006 | A1 |
20060101140 | Gai et al. | May 2006 | A1 |
20060104218 | Kako | May 2006 | A1 |
20060146832 | Rampal et al. | Jul 2006 | A1 |
20060171318 | Bergamasco et al. | Aug 2006 | A1 |
20060187832 | Yu | Aug 2006 | A1 |
20060187900 | Akbar | Aug 2006 | A1 |
20060198323 | Finn | Sep 2006 | A1 |
20060215550 | Malhotra | Sep 2006 | A1 |
20060251067 | DeSanti et al. | Nov 2006 | A1 |
20060274791 | Garcia et al. | Dec 2006 | A1 |
20070041321 | Sasaki et al. | Feb 2007 | A1 |
20070047443 | Desai et al. | Mar 2007 | A1 |
20070081454 | Bergamasco et al. | Apr 2007 | A1 |
20070115824 | Chandra et al. | May 2007 | A1 |
20070121617 | Kanekar et al. | May 2007 | A1 |
20070183332 | Oh et al. | Aug 2007 | A1 |
20080019282 | Alaria et al. | Jan 2008 | A1 |
20080069114 | Shimada et al. | Mar 2008 | A1 |
20080089247 | Sane et al. | Apr 2008 | A1 |
20080117938 | Erich et al. | May 2008 | A1 |
20080186968 | Farinacci et al. | Aug 2008 | A1 |
20080212595 | Figueira et al. | Sep 2008 | A1 |
20080259798 | Loh et al. | Oct 2008 | A1 |
20080273465 | Gusat et al. | Nov 2008 | A1 |
20090052326 | Bargamasco et al. | Feb 2009 | A1 |
20090073882 | McAlpine et al. | Mar 2009 | A1 |
20090232138 | Gobara et al. | Sep 2009 | A1 |
20090252038 | Cafiero et al. | Oct 2009 | A1 |
Number | Date | Country |
---|---|---|
1778079 | May 2006 | CN |
1206099 | May 2002 | EP |
WO2004064324 | Jul 2004 | WO |
WO 2006047092 | May 2006 | WO |
WO 2006047109 | May 2006 | WO |
WO 2006047194 | May 2006 | WO |
WO 2006047223 | May 2006 | WO |
WO 2006057730 | Jun 2006 | WO |
WO 2006063922 | Jun 2006 | WO |
WO 2007050250 | May 2007 | WO |
Number | Date | Country | |
---|---|---|---|
20090010162 A1 | Jan 2009 | US |