DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-10 rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 1 recites the limitation "the M threads" in lines 6-7. There is insufficient antecedent basis for this limitation in the claim.
Claim 1 recites the limitation "the M threads" in lines 10-11. There is insufficient antecedent basis for this limitation in the claim.
Claim 1 recites the limitation "the M threads" in lines 15-16. There is insufficient antecedent basis for this limitation in the claim.
Claim 1 has been amended to recite “up to M threads per processing core”, as such, it is unclear what “the M threads” refers to, as each processing core has up to M threads.
Claims not referred to specifically above are rejected as depending from a rejected claim.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim(s) 1-8 and 11, 12, 14-18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Bozek et al. (Pub. No. US 2012/0226866) in view of LeMay et al. (Pub. No. US 2016/0092673).
Claim 1:
Bozek et al. disclose a method of reducing cache thrashing in a processing system, comprising:
issuing a workload [par. 0081 – “Embodiments of the invention track virtual machine application workloads, and their processor cache hit ratios. Then based on the cache hit ratios, use dynamic migration to move virtual machine workloads with a higher processor cache demand to physical servers with larger L1, L2 and/or L3 caches. Conversely, embodiments may identify virtual machine workloads with a lower cache appetite and migrate those virtual machine workloads to physical servers with more moderate L1, L2 and/or L3 cache sizes. Using this methodology results in increased utilization of physical servers and increased performance of cache-sensitive virtual machine workloads. More specifically, the cache demands of a virtual machine workload, in the form of cache hit ratio data, are obtained from the processors and provided to the provisioning manager 222 and ultimately to the global provisioning manager 232 (See FIG. 5). The global provisioning manager migrates high cache demand workloads to physical servers with larger cache.”];
transmitting a memory access request associated with the workload to a first-level cache of the processing system and determining whether a first cache hit or a first cache miss occurs at the first-level cache [pars. 0006, 0020-0021, 0080 – A request is sent to the L1 cache. (“If an instruction can access either Level 1, 2, or 3 cache, this operation results in a cache hit. Otherwise, the processor must go to external memory (DIMMs 264) to obtain the data resulting in a much longer instruction time. It is most efficient if instructions are able to use information in the Level 1 cache so that the data can be accessed immediately.”)];
transmitting the memory access request associated with the workload to a second-level cache of the processing system in response to the first cache miss and determining whether a second cache hit or a second cache miss occurs at the second-level cache [pars. 0006, 0020-0021, 0080 – The request is sent to the L2 cache when it misses in the L1 cache. (“If an instruction can access either Level 1, 2, or 3 cache, this operation results in a cache hit. Otherwise, the processor must go to external memory (DIMMs 264) to obtain the data resulting in a much longer instruction time. It is most efficient if instructions are able to use information in the Level 1 cache so that the data can be accessed immediately.”)];
transmitting the memory access request associated with the workload to a main memory of the processing system in response to the second cache miss [pars. 0006, 0020-0021, 0080 – The request is sent to main memory when it misses in the cache hierarchy. (“If an instruction can access either Level 1, 2, or 3 cache, this operation results in a cache hit. Otherwise, the processor must go to external memory (DIMMs 264) to obtain the data resulting in a much longer instruction time. It is most efficient if instructions are able to use information in the Level 1 cache so that the data can be accessed immediately.”)];
determining whether a relationship between a first hit rate of the first-level cache and a second hit rate of the second-level cache satisfies a predetermined criterion [par. 0091 – “For example, in Physical Server #1, the first virtual machine VM1 has an L1 cache hit ratio equal to the number of L1 cache hits (i.e., 800) divided by the total number of memory accesses (i.e., 1000). Accordingly, the L1 cache hit ratio is 0.80 (alternatively expressed as 80 percent). Since there were only 200 L1 misses (i.e., 1000 memory accesses minus 800 L1 cache hits), there are only 200 potential memory accesses to the L2 cache. The L2 cache hit ratio is, therefore, equal to the number of L2 cache hits (i.e., 160) divided by the 200 memory accesses to the L2 cache. Accordingly the L2 cache hit ratio is 0.80. Similarly, the L3 cache hit ratio of 0.25 is calculated by dividing the 10 L3 cache hits divided by the 40 L2 cache misses. By comparing the class-specific cache hit ratios with the respective threshold ratios, it is seen that the L1 and L2 cache hit ratios are greater than their respective threshold ratios, but the L3 cache hit ratio is less than the respective threshold ratio. In other words, there is an L3 cache hit ratio exception. As a result, the virtual machine VM1 is identified as a candidate for migration to another physical server. The selection of an appropriate target physical server is discussed in relation to FIG. 8.”]; and
decreasing a value of the workload when the predetermined criterion is satisfied [par. 0091 – The VM is migrated (e.g. workload is set to zero on the previous machine). (“For example, in Physical Server #1, the first virtual machine VM1 has an L1 cache hit ratio equal to the number of L1 cache hits (i.e., 800) divided by the total number of memory accesses (i.e., 1000). Accordingly, the L1 cache hit ratio is 0.80 (alternatively expressed as 80 percent). Since there were only 200 L1 misses (i.e., 1000 memory accesses minus 800 L1 cache hits), there are only 200 potential memory accesses to the L2 cache. The L2 cache hit ratio is, therefore, equal to the number of L2 cache hits (i.e., 160) divided by the 200 memory accesses to the L2 cache. Accordingly the L2 cache hit ratio is 0.80. Similarly, the L3 cache hit ratio of 0.25 is calculated by dividing the 10 L3 cache hits divided by the 40 L2 cache misses. By comparing the class-specific cache hit ratios with the respective threshold ratios, it is seen that the L1 and L2 cache hit ratios are greater than their respective threshold ratios, but the L3 cache hit ratio is less than the respective threshold ratio. In other words, there is an L3 cache hit ratio exception. As a result, the virtual machine VM1 is identified as a candidate for migration to another physical server. The selection of an appropriate target physical server is discussed in relation to FIG. 8.”], wherein:
a storage capacity of the second-level cache is higher than a storage capacity of the first-level cache [fig. 6; par. 0080 – “FIG. 6 also shows the cache that is available to the processor in the server 240. The Level 1 cache 251 is local to the processor core 254 and can be accessed the fastest. The Level 2 cache 252 is still located on the processor module 250, but is further away from the core processor and although larger than the size of Level 1 cache is accessed slower.”];
an access latency of the second-level cache is higher than an access latency of the first-level cache [fig. 6; par. 0080 – “FIG. 6 also shows the cache that is available to the processor in the server 240. The Level 1 cache 251 is local to the processor core 254 and can be accessed the fastest. The Level 2 cache 252 is still located on the processor module 250, but is further away from the core processor and although larger than the size of Level 1 cache is accessed slower.”];
However, Bozek et al. do not specifically disclose,
issuing, by a scheduler to a plurality of processing cores of the processing system, up to M threads per processing core to process a workload;
M is an integer larger than 1.
In the same field of endeavor, LeMay et al. disclose,
issuing, by a scheduler to a plurality of processing cores of the processing system, up to M threads per processing core to process a workload [pars. 0021, 0027-0028 – “The processor 120 may be embodied as any type of processor capable of performing the functions described herein. For example, the processor 120 may be embodied as a single or multi-core processor(s), digital signal processor, microcontroller, or other processor or processing/controlling circuit.” … “Although illustrated as including a single guest virtual machine 206 with a single software thread 208, it should be understood that the environment 200 may include many guest virtual machines 206, and each guest virtual machine may include many software threads 208.” … “The hypervisor 212 may be embodied as any virtual machine monitor, hypervisor, or other component that allows a virtualized workload to be executed on the computing device 100. The hypervisor 212 may have complete control over the computing device 100, for example by executing in a VMX-root mode. The hypervisor 212 is configured to protect part of the memory 126 from access by the guest virtual machine 206. The hypervisor establishes a shadow stack module 214, an out-of-bounds module 220, and a mismatch module 224. Although illustrated as a hypervisor 212, it should be understood that the environment 200 may instead include another privileged component that does not execute virtualized workloads, such as an operating system executing in ringlevel 0. Those embodiments may provide some level of protection against ROP exploits, even if the shadow stack may be stored in memory 126 that is accessible to the software being protected (e.g., the software thread 208).”];
M is an integer larger than 1 [par. 0027 – “Although illustrated as including a single guest virtual machine 206 with a single software thread 208, it should be understood that the environment 200 may include many guest virtual machines 206, and each guest virtual machine may include many software threads 208.”].
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Bozek et al. to include multithreading, as taught by LeMay et al., in order to improve performance.
Claim 2 (as applied to claim 1 above):
Bozek et al. disclose, wherein:
the predetermined criterion is satisfied when the second hit rate is higher than the first hit rate [par. 0091 – The criteria is met when any cache fails to meet the respective threshold ratio. In such a case, the L2 hit rate may be higher than the first hit rate. (“For example, in Physical Server #1, the first virtual machine VM1 has an L1 cache hit ratio equal to the number of L1 cache hits (i.e., 800) divided by the total number of memory accesses (i.e., 1000). Accordingly, the L1 cache hit ratio is 0.80 (alternatively expressed as 80 percent). Since there were only 200 L1 misses (i.e., 1000 memory accesses minus 800 L1 cache hits), there are only 200 potential memory accesses to the L2 cache. The L2 cache hit ratio is, therefore, equal to the number of L2 cache hits (i.e., 160) divided by the 200 memory accesses to the L2 cache. Accordingly the L2 cache hit ratio is 0.80. Similarly, the L3 cache hit ratio of 0.25 is calculated by dividing the 10 L3 cache hits divided by the 40 L2 cache misses. By comparing the class-specific cache hit ratios with the respective threshold ratios, it is seen that the L1 and L2 cache hit ratios are greater than their respective threshold ratios, but the L3 cache hit ratio is less than the respective threshold ratio. In other words, there is an L3 cache hit ratio exception. As a result, the virtual machine VM1 is identified as a candidate for migration to another physical server. The selection of an appropriate target physical server is discussed in relation to FIG. 8.”)].
Claim 3 (as applied to claim 1 above):
Bozek et al. disclose, wherein:
the predetermined criterion is satisfied when the second hit rate is higher than the first hit rate by more than a predetermined value [par. 0091 – The criteria is met when any cache fails to meet the respective threshold ratio. In such a case, the L2 hit rate may be higher than the L1 hit rate by a value greater than any of the respective hit rate thresholds (e.g. a predetermined value). (“For example, in Physical Server #1, the first virtual machine VM1 has an L1 cache hit ratio equal to the number of L1 cache hits (i.e., 800) divided by the total number of memory accesses (i.e., 1000). Accordingly, the L1 cache hit ratio is 0.80 (alternatively expressed as 80 percent). Since there were only 200 L1 misses (i.e., 1000 memory accesses minus 800 L1 cache hits), there are only 200 potential memory accesses to the L2 cache. The L2 cache hit ratio is, therefore, equal to the number of L2 cache hits (i.e., 160) divided by the 200 memory accesses to the L2 cache. Accordingly the L2 cache hit ratio is 0.80. Similarly, the L3 cache hit ratio of 0.25 is calculated by dividing the 10 L3 cache hits divided by the 40 L2 cache misses. By comparing the class-specific cache hit ratios with the respective threshold ratios, it is seen that the L1 and L2 cache hit ratios are greater than their respective threshold ratios, but the L3 cache hit ratio is less than the respective threshold ratio. In other words, there is an L3 cache hit ratio exception. As a result, the virtual machine VM1 is identified as a candidate for migration to another physical server. The selection of an appropriate target physical server is discussed in relation to FIG. 8.”)].
Claim 4 (as applied to claim 1 above):
Bozek et al. disclose, wherein:
the predetermined criterion is satisfied when the second hit rate is higher than a predetermined positive value [par. 0091 – The criteria is met when any cache fails to meet the respective threshold ratio. In such a case, the L2 cache may still meet the respective threshold value. (“For example, in Physical Server #1, the first virtual machine VM1 has an L1 cache hit ratio equal to the number of L1 cache hits (i.e., 800) divided by the total number of memory accesses (i.e., 1000). Accordingly, the L1 cache hit ratio is 0.80 (alternatively expressed as 80 percent). Since there were only 200 L1 misses (i.e., 1000 memory accesses minus 800 L1 cache hits), there are only 200 potential memory accesses to the L2 cache. The L2 cache hit ratio is, therefore, equal to the number of L2 cache hits (i.e., 160) divided by the 200 memory accesses to the L2 cache. Accordingly the L2 cache hit ratio is 0.80. Similarly, the L3 cache hit ratio of 0.25 is calculated by dividing the 10 L3 cache hits divided by the 40 L2 cache misses. By comparing the class-specific cache hit ratios with the respective threshold ratios, it is seen that the L1 and L2 cache hit ratios are greater than their respective threshold ratios, but the L3 cache hit ratio is less than the respective threshold ratio. In other words, there is an L3 cache hit ratio exception. As a result, the virtual machine VM1 is identified as a candidate for migration to another physical server. The selection of an appropriate target physical server is discussed in relation to FIG. 8.”)].
Claim 5 (as applied to claim 1 above):
Bozek et al. disclose, wherein:
a storage capacity of the main memory is higher than the storage capacity of the second-level cache [fig. par. 0080 – “FIG. 6 also shows the cache that is available to the processor in the server 240. The Level 1 cache 251 is local to the processor core 254 and can be accessed the fastest. The Level 2 cache 252 is still located on the processor module 250, but is further away from the core processor and although larger than the size of Level 1 cache is accessed slower. The Level 3 cache 253 is located off the processor module 250 and has more capacity, albeit at a slower access rate, than the Level 2 cache. If an instruction can access either Level 1, 2, or 3 cache, this operation results in a cache hit. Otherwise, the processor must go to external memory (DIMMs 264) to obtain the data resulting in a much longer instruction time. It is most efficient if instructions are able to use information in the Level 1 cache so that the data can be accessed immediately.”];
an access latency of the main memory is higher than the access latency of the second-level cache [fig. par. 0080 – “FIG. 6 also shows the cache that is available to the processor in the server 240. The Level 1 cache 251 is local to the processor core 254 and can be accessed the fastest. The Level 2 cache 252 is still located on the processor module 250, but is further away from the core processor and although larger than the size of Level 1 cache is accessed slower. The Level 3 cache 253 is located off the processor module 250 and has more capacity, albeit at a slower access rate, than the Level 2 cache. If an instruction can access either Level 1, 2, or 3 cache, this operation results in a cache hit. Otherwise, the processor must go to external memory (DIMMs 264) to obtain the data resulting in a much longer instruction time. It is most efficient if instructions are able to use information in the Level 1 cache so that the data can be accessed immediately.”].
Claim 6 (as applied to claim 1 above):
Bozek et al. disclose the method, further comprising:
providing, by the first-level cache, data requested by the memory access request for completing an operation associated with the memory access request in response to the first cache hit [pars. 0006, 0020-0021, 0080 – Data is returned from the level 1 cache in response to a cache hit. (“If an instruction can access either Level 1, 2, or 3 cache, this operation results in a cache hit. Otherwise, the processor must go to external memory (DIMMs 264) to obtain the data resulting in a much longer instruction time. It is most efficient if instructions are able to use information in the Level 1 cache so that the data can be accessed immediately.”)].
Claim 7 (as applied to claim 1 above):
Bozek et al. disclose the method, further comprising:
providing, by the second-level cache, data requested by the memory access request for completing an operation associated with the memory access request in response to the second cache hit [pars. 0006, 0020-0021, 0080 – Data is returned from the level 2 cache in response to a cache hit. (“If an instruction can access either Level 1, 2, or 3 cache, this operation results in a cache hit. Otherwise, the processor must go to external memory (DIMMs 264) to obtain the data resulting in a much longer instruction time. It is most efficient if instructions are able to use information in the Level 1 cache so that the data can be accessed immediately.”)].
Claim 8 (as applied to claim 1 above):
Bozek et al. disclose the method, further comprising:
providing, by the main memory, data requested by the memory access request for completing an operation associated with the memory access request in response to the first cache miss and the second cache miss [pars. 0006, 0020-0021, 0080 – Data is returned from the main memory in response to a miss in the cache hierarchy. (“If an instruction can access either Level 1, 2, or 3 cache, this operation results in a cache hit. Otherwise, the processor must go to external memory (DIMMs 264) to obtain the data resulting in a much longer instruction time. It is most efficient if instructions are able to use information in the Level 1 cache so that the data can be accessed immediately.”)].
Claim 11:
Bozek et al. disclose a processing system which reduces cache thrashing, comprising:
a first-level cache with a first storage capacity and a first access latency [fig. 6; par. 0080 – “FIG. 6 also shows the cache that is available to the processor in the server 240. The Level 1 cache 251 is local to the processor core 254 and can be accessed the fastest. The Level 2 cache 252 is still located on the processor module 250, but is further away from the core processor and although larger than the size of Level 1 cache is accessed slower.”], and configured to:
receive a memory access request associated with a workload [pars. 0006, 0020-0021, 0080 – A request is sent to the L1 cache. (“If an instruction can access either Level 1, 2, or 3 cache, this operation results in a cache hit. Otherwise, the processor must go to external memory (DIMMs 264) to obtain the data resulting in a much longer instruction time. It is most efficient if instructions are able to use information in the Level 1 cache so that the data can be accessed immediately.”)]; and
determine whether a first cache hit or a first cache miss occurs at the first-level cache [pars. 0006, 0020-0021, 0080 – It is determined whether a hit or miss occurs. (“If an instruction can access either Level 1, 2, or 3 cache, this operation results in a cache hit. Otherwise, the processor must go to external memory (DIMMs 264) to obtain the data resulting in a much longer instruction time. It is most efficient if instructions are able to use information in the Level 1 cache so that the data can be accessed immediately.”)];
a second-level cache with a second storage capacity and a second access latency [fig. 6; par. 0080 – “FIG. 6 also shows the cache that is available to the processor in the server 240. The Level 1 cache 251 is local to the processor core 254 and can be accessed the fastest. The Level 2 cache 252 is still located on the processor module 250, but is further away from the core processor and although larger than the size of Level 1 cache is accessed slower.”], and configured to:
receive the memory access request associated with the workload from the first-level cache in response to a first cache miss at the first-level cache [pars. 0006, 0020-0021, 0080 – The request is sent to the L2 cache when it misses in the L1 cache. (“If an instruction can access either Level 1, 2, or 3 cache, this operation results in a cache hit. Otherwise, the processor must go to external memory (DIMMs 264) to obtain the data resulting in a much longer instruction time. It is most efficient if instructions are able to use information in the Level 1 cache so that the data can be accessed immediately.”)]; and
determine whether a second cache hit or a second cache miss occurs at the second-level cache [pars. 0006, 0020-0021, 0080 – It is determined whether a hit or miss occurs. (“If an instruction can access either Level 1, 2, or 3 cache, this operation results in a cache hit. Otherwise, the processor must go to external memory (DIMMs 264) to obtain the data resulting in a much longer instruction time. It is most efficient if instructions are able to use information in the Level 1 cache so that the data can be accessed immediately.”)]; and
a main memory configured to receive the memory access request associated with the workload from the second-level cache in response to a second cache miss at the second-level cache [pars. 0006, 0020-0021, 0080 – The request is sent to main memory when it misses in the cache hierarchy. (“If an instruction can access either Level 1, 2, or 3 cache, this operation results in a cache hit. Otherwise, the processor must go to external memory (DIMMs 264) to obtain the data resulting in a much longer instruction time. It is most efficient if instructions are able to use information in the Level 1 cache so that the data can be accessed immediately.”)]; and
a scheduler configured to:
transmit the memory access request associated with the workload to the first-level cache [pars. 0006, 0020-0021, 0080 – A request is sent to the L1 cache. (“If an instruction can access either Level 1, 2, or 3 cache, this operation results in a cache hit. Otherwise, the processor must go to external memory (DIMMs 264) to obtain the data resulting in a much longer instruction time. It is most efficient if instructions are able to use information in the Level 1 cache so that the data can be accessed immediately.”)];
determine whether a relationship between a first hit rate of the first-level cache and a second hit rate of the second-level cache satisfies a predetermined criterion [par. 0091 – “For example, in Physical Server #1, the first virtual machine VM1 has an L1 cache hit ratio equal to the number of L1 cache hits (i.e., 800) divided by the total number of memory accesses (i.e., 1000). Accordingly, the L1 cache hit ratio is 0.80 (alternatively expressed as 80 percent). Since there were only 200 L1 misses (i.e., 1000 memory accesses minus 800 L1 cache hits), there are only 200 potential memory accesses to the L2 cache. The L2 cache hit ratio is, therefore, equal to the number of L2 cache hits (i.e., 160) divided by the 200 memory accesses to the L2 cache. Accordingly the L2 cache hit ratio is 0.80. Similarly, the L3 cache hit ratio of 0.25 is calculated by dividing the 10 L3 cache hits divided by the 40 L2 cache misses. By comparing the class-specific cache hit ratios with the respective threshold ratios, it is seen that the L1 and L2 cache hit ratios are greater than their respective threshold ratios, but the L3 cache hit ratio is less than the respective threshold ratio. In other words, there is an L3 cache hit ratio exception. As a result, the virtual machine VM1 is identified as a candidate for migration to another physical server. The selection of an appropriate target physical server is discussed in relation to FIG. 8.”]; and
decrease a value of the workload when the predetermined criterion is satisfied [par. 0091 – The VM is migrated (e.g. workload is set to zero on the previous machine). (“For example, in Physical Server #1, the first virtual machine VM1 has an L1 cache hit ratio equal to the number of L1 cache hits (i.e., 800) divided by the total number of memory accesses (i.e., 1000). Accordingly, the L1 cache hit ratio is 0.80 (alternatively expressed as 80 percent). Since there were only 200 L1 misses (i.e., 1000 memory accesses minus 800 L1 cache hits), there are only 200 potential memory accesses to the L2 cache. The L2 cache hit ratio is, therefore, equal to the number of L2 cache hits (i.e., 160) divided by the 200 memory accesses to the L2 cache. Accordingly the L2 cache hit ratio is 0.80. Similarly, the L3 cache hit ratio of 0.25 is calculated by dividing the 10 L3 cache hits divided by the 40 L2 cache misses. By comparing the class-specific cache hit ratios with the respective threshold ratios, it is seen that the L1 and L2 cache hit ratios are greater than their respective threshold ratios, but the L3 cache hit ratio is less than the respective threshold ratio. In other words, there is an L3 cache hit ratio exception. As a result, the virtual machine VM1 is identified as a candidate for migration to another physical server. The selection of an appropriate target physical server is discussed in relation to FIG. 8.”], wherein:
the second storage capacity is higher than the first storage capacity [fig. 6; par. 0080 – “FIG. 6 also shows the cache that is available to the processor in the server 240. The Level 1 cache 251 is local to the processor core 254 and can be accessed the fastest. The Level 2 cache 252 is still located on the processor module 250, but is further away from the core processor and although larger than the size of Level 1 cache is accessed slower.”];
the second access latency is higher than the first access latency [fig. 6; par. 0080 – “FIG. 6 also shows the cache that is available to the processor in the server 240. The Level 1 cache 251 is local to the processor core 254 and can be accessed the fastest. The Level 2 cache 252 is still located on the processor module 250, but is further away from the core processor and although larger than the size of Level 1 cache is accessed slower.”];
However, Bozek et al. do not specifically disclose,
a plurality of processing cores;
the scheduler configured to:
issue the M threads per processing core to each of the plurality of processing cores for processing a workload;
M is an integer larger than 1.
In the same field of endeavor, LeMay et al. disclose,
a plurality of processing cores [par. 0021 – “The processor 120 may be embodied as any type of processor capable of performing the functions described herein. For example, the processor 120 may be embodied as a single or multi-core processor(s), digital signal processor, microcontroller, or other processor or processing/controlling circuit.”];
the scheduler configured to:
issue the M threads per processing core to each of the plurality of processing cores for processing a workload [pars. 0021, 0027-0028 – “The processor 120 may be embodied as any type of processor capable of performing the functions described herein. For example, the processor 120 may be embodied as a single or multi-core processor(s), digital signal processor, microcontroller, or other processor or processing/controlling circuit.” … “Although illustrated as including a single guest virtual machine 206 with a single software thread 208, it should be understood that the environment 200 may include many guest virtual machines 206, and each guest virtual machine may include many software threads 208.” … “The hypervisor 212 may be embodied as any virtual machine monitor, hypervisor, or other component that allows a virtualized workload to be executed on the computing device 100. The hypervisor 212 may have complete control over the computing device 100, for example by executing in a VMX-root mode. The hypervisor 212 is configured to protect part of the memory 126 from access by the guest virtual machine 206. The hypervisor establishes a shadow stack module 214, an out-of-bounds module 220, and a mismatch module 224. Although illustrated as a hypervisor 212, it should be understood that the environment 200 may instead include another privileged component that does not execute virtualized workloads, such as an operating system executing in ringlevel 0. Those embodiments may provide some level of protection against ROP exploits, even if the shadow stack may be stored in memory 126 that is accessible to the software being protected (e.g., the software thread 208).”];
M is an integer larger than 1 [par. 0027 – “Although illustrated as including a single guest virtual machine 206 with a single software thread 208, it should be understood that the environment 200 may include many guest virtual machines 206, and each guest virtual machine may include many software threads 208.”].
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Bozek et al. to include multithreading, as taught by LeMay et al., in order to improve performance.
Claim 12 (as applied to claim 11 above):
Claim 12, directed to a processing system, is rejected for the same reasons set forth in the rejection of claim 2 above, mutatis mutandis.
Claim 14 (as applied to claim 11 above):
Claim 14, directed to a processing system, is rejected for the same reasons set forth in the rejection of claim 4 above, mutatis mutandis.
Claim 15 (as applied to claim 11 above):
Claim 15, directed to a processing system, is rejected for the same reasons set forth in the rejection of claim 5 above, mutatis mutandis.
Claim 16 (as applied to claim 11 above):
Claim 16, directed to a processing system, is rejected for the same reasons set forth in the rejection of claim 6 above, mutatis mutandis.
Claim 17 (as applied to claim 11 above):
Claim 17, directed to a processing system, is rejected for the same reasons set forth in the rejection of claim 7 above, mutatis mutandis.
Claim 18 (as applied to claim 11 above):
Claim 18, directed to a processing system, is rejected for the same reasons set forth in the rejection of claim 8 above, mutatis mutandis.
Claim(s) 9, 10, and 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over Bozek et al. (Pub. No. US 2012/0226866) in view of LeMay et al. (Pub. No. US 2016/0092673) as applied to claims 1 and 11 above, respectively, and further in view of Zheng et al. (Pub. No. US 2020/0065126).
Claim 9 (as applied to claim 1 above):
Bozek et al. and LeMay et al. disclose all the limitations above but do not specifically disclose the method, further comprising:
determining whether an adjustment made to the value of M has met a predetermined condition; and
increasing the value of M when the predetermined condition is met.
In the same field of endeavor, Zheng et al. disclose,
determining whether an adjustment made to the value of M has met a predetermined condition [par. 0067 – “At step 513, the resource management service 246 can check the last migration 242 to determine when the virtual machine 106 was last migrated from one host to another. To avoid the performance impacts of thrashing, where a VM 106 is repeatedly migrated between hosts in quick succession, the resource management service 246 may weigh the determination of whether or not to cause the VM 106 to migrate again. Accordingly, the resource management service 246 can then determine whether the time that the virtual machine 106 was last migrated occurred within a predefined window of time prior to the current time. If the VM 106 was last migrated outside of the predefined window of time, then the process skips to step 519. However, if the VM's 106 last migration 242 occurred within the predefined window of time, then the process proceeds to step 516.”]; and
increasing the value of M when the predetermined condition is met [par. 0067 – “At step 513, the resource management service 246 can check the last migration 242 to determine when the virtual machine 106 was last migrated from one host to another. To avoid the performance impacts of thrashing, where a VM 106 is repeatedly migrated between hosts in quick succession, the resource management service 246 may weigh the determination of whether or not to cause the VM 106 to migrate again. Accordingly, the resource management service 246 can then determine whether the time that the virtual machine 106 was last migrated occurred within a predefined window of time prior to the current time. If the VM 106 was last migrated outside of the predefined window of time, then the process skips to step 519. However, if the VM's 106 last migration 242 occurred within the predefined window of time, then the process proceeds to step 516.”].
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the combined teachings of Bozek et al. and LeMay et al. to include evaluating a last migration time, as taught by Zheng et al., in order to improve performance by reducing thrashing.
Claim 10 (as applied to claim 9 above):
Zheng et al. disclose, wherein the predetermined condition is met when:
the value of M has been reduced more than K times; a difference between an original value of M and a current value of N exceeds a predetermined value; or a predetermined period of time has elapsed since a first decrease of the value of M [par. 0067 – “At step 513, the resource management service 246 can check the last migration 242 to determine when the virtual machine 106 was last migrated from one host to another. To avoid the performance impacts of thrashing, where a VM 106 is repeatedly migrated between hosts in quick succession, the resource management service 246 may weigh the determination of whether or not to cause the VM 106 to migrate again. Accordingly, the resource management service 246 can then determine whether the time that the virtual machine 106 was last migrated occurred within a predefined window of time prior to the current time. If the VM 106 was last migrated outside of the predefined window of time, then the process skips to step 519. However, if the VM's 106 last migration 242 occurred within the predefined window of time, then the process proceeds to step 516.”].
Claim 19 (as applied to claim 11 above):
Claim 19, directed to a processing system, is rejected for the same reasons set forth in the rejection of claim 9 above, mutatis mutandis.
Claim(s) 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Bozek et al. (Pub. No. US 2012/0226866) in view of LeMay et al. (Pub. No. US 2016/0092673) as applied to claim 11 above, and further in view of Tuan (Pub. No. US 2013/0226535).
Claim 20 (as applied to claim 11 above):
Bozek et al. and LeMay et al. disclose all the limitations above but do not specifically disclose,
wherein the plurality of processing cores, the scheduler and the first-level cache are implemented as a streaming multiprocessor (SM).
In the same field of endeavor, Tuan discloses,
wherein the plurality of processing cores, the scheduler and the first-level cache are implemented as a streaming multiprocessor (SM) [fig. 2; par. 0054 – “FIG. 2 shows GPU architecture 200 that is suitable for implementing the concurrent simulation system of the present invention. GPU 200 may be, for example, a GPU having the Nvidia Fermi architecture. As shown in FIG. 2, GPU architecture 200 includes an integrated circuit having 16 streaming multiprocessors (e.g., streaming multiprocessor (SM) 201), second-level or "L2" cache 206, and six on-chip dynamic random access memory (DRAM) controllers 207-1 to 207-6 (collectively, "DRAM controllers 207") for accessing a global memory shared by all the SMs. DRAM controllers 207 may be, for example, configured to access a 64-bit wide memory. Each streaming processor includes 32 processor cores (e.g., processor core 202), register file 203, which includes 4096 32-bit registers, 64K-byte of memory divided between shared memory 206 and first-level (i.e., "L1") cache 204, and control circuitry including an instruction cache, warp schedulers and dispatch units. In one implementation, the 64K-byte memory may be divided into either as a 16K-byte shared memory 206 and 48K-byte cache.”].
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the combined teachings of Bozek et al. and LeMay et al. to include a streaming processor, as taught by Tuan, in order to improve performance and reduce power consumption.
Allowable Subject Matter
Claim 13 objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
Response to Arguments
Applicant's arguments filed May 11, 2026 have been fully considered but they are not persuasive.
Applicant argues that “decreasing a value of M” is not satisfied by setting to zero.
Examiner submits that decreasing, by definition, is to make or become smaller, fewer, or less in size, amount, number, or strength. Setting a value of something to zero decreases its value.
In response to applicant's argument that the references fail to show certain features of the invention, it is noted that the features upon which applicant relies (i.e., inter-system resource migration) are not recited in the rejected claim(s). Although the claims are interpreted in light of the specification, limitations from the specification are not read into the claims. See In re Van Geuns, 988 F.2d 1181, 26 USPQ2d 1057 (Fed. Cir. 1993).
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to LARRY T MACKALL whose telephone number is (571)270-1172. The examiner can normally be reached Monday - Friday, 9am-5pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Reginald G Bragdon can be reached at (571) 272-4204. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
LARRY T. MACKALL
Primary Examiner
Art Unit 2131
5 August 2026
/LARRY T MACKALL/Primary Examiner, Art Unit 2139
Prosecution Insights