Prosecution Insights
Last updated: November 16, 2026
Application No. 19/044,612

FAST MODULO PROCESSING FOR INTERLEAVED MEMORY HAVING NON-POWER-OF-TWO CHANNELS

Final Rejection §103§112
Filed
Feb 03, 2025
Examiner
CHAPPELL, DANIEL C
Art Unit
2135
Tech Center
2100 — Computer Architecture & Software
Assignee
Arteris Inc.
OA Round
2 (Final)
81%
Grant Probability
Favorable
3-4
OA Rounds
99%
With Interview

Examiner Intelligence

Grants 81% — above average
81%
Career Allowance Rate
500 granted / 618 resolved
+25.9% vs TC avg
Strong +45% interview lift
Without
With
+45.1%
Interview Lift
resolved cases with interview
Typical timeline
Avg Prosecution
17 currently pending
Career history
641
Total Applications
across all art units

Office Action

§103 §112
DETAILED ACTION 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 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 Office action is in response to communications dated 6/24/2026. Claims 1-4, 8-10, 15, and 17-20 are amended. Claims 1-20 are pending. Claims 1-20 are rejected. The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action. Claim Rejections - 35 USC § 112 The Examiner thanks Applicant for amending the claims to cure the rejections of claims 1-20 under 35 U.S.C. §112(b) and therefore respectfully withdraws the rejections of claims 1-20 under 35 U.S.C. §112(b) made in the non-final Office action dated 2/24/2026. Claim Rejections - 35 USC § 103 Claims 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over non-patent literature “Memory Interleaving” (“Matloff”) in view of USPGPUB 2016/0357666 (“Avoinne”) in view of USPGPUB 2004/0139290 (“Wolrich”) and further in view of USPGPUB 2015/0089168 (“Kalyanasundharam”). As per claim 1, Matloff substantially teaches a hardware-implemented method for an electronic system including system memory having N interleaved channels, where N is not a power of two, the method comprising: a hardware-implemented method for an electronic system including system memory having N interleaved channels, where N is not a power of two: (Matloff, page 1, section 1 “Introduction,” paragraphs 1-3; and page 5, section “3.2 Prime Number of Modules,” where the computer system of Matloff comprises an interleaved memory system that utilizes a prime number A of memory modules (i.e., memory channels) that is not a power of 2. Matloff therefore substantially teaches a hardware-implemented method for an electronic system including system memory having N interleaved channels, where N is not a power of two); receiving a request to access the system memory, the request including an M-bit address: (Matloff, page 1, section 1 “Introduction,” paragraphs 1-3; and page 5, section “3.2 Prime Number of Modules,” where the computer system of Matloff responds to a request for requested words of memory and thus must have received a memory request. The Examiner notes that the example memory architecture of Matloff uses power-of-two addressing which means that a memory address will comprise M bits. Matloff therefore substantially teaches receiving a request to access the system memory, the request including an M-bit address). Matloff does not appear to explicitly teach the other limitations of this claim beyond those taught above; however, in an analogous art, Avoinne teaches memory interleaving on memory channels. As per claim 1, Avoinne particularly teaches: and processing the address to calculate a modulo N of the address without performing division-by-N, wherein the modulo N of the address is calculated as [∑iaimod N * 2bi mod N] mod N, where ∑iai * 2bi represents the M-bit address, ai is an ith word in the address, 2bi is a power of two associated with the ith word, and modulo N of 2bi = 1 for all i: (Avoinne, Abstract; FIG. 5; and paragraphs 0006 and 0026-0028, where the system of Avoinne merely examines received bits of an address in order to determine a channel to which the address should be routed. Although Avoinne does not appear to explicitly teach using the claimed equation of [∑iaimod N * 2bi mod N] mod N, where ∑iai * 2bi represents the M-bit address, ai is an ith word in the address, 2bi is a power of two associated with the ith word, and modulo N of 2bi = 1 for all i in order to calculate the modulo N of the address, it would have been obvious to a person having ordinary skill in the art to use the claimed equation to calculate the modulo N of the address. The modification would have been obvious because a person having ordinary skill in the art before the instant application was effectively filed would be motivated to increase system flexibility and performance by enabling usage of non-power-of-2 interleaving using computer processor hardware to perform binary operations to calculate addresses. Avoinne therefore particularly teaches and processing the address to calculate a modulo N of the address without performing division-by-N, wherein the modulo N of the address is calculated as [∑iaimod N * 2bi mod N] mod N, where ∑iai * 2bi represents the M-bit address, ai is an ith word in the address, 2bi is a power of two associated with the ith word, and modulo N of 2bi = 1 for all i). It would have been obvious to a person having ordinary skill in the art, having the teachings of Avoinne and Matloff before them before the instant application was effectively filed, to modify the system of Matloff to use the principles of Avoinne of hardware memory interleaving. The modification would have been obvious because a person having ordinary skill in the art would be motivated to increase system performance by implementing techniques to perform memory interleaving in hardware. Neither Matloff nor Avoinne appears to explicitly teach the other limitations of this claim beyond those taught above; however, in an analogous art, Wolrich teaches memory interleaving. As per claim 1, Wolrich particularly teaches: wherein in a first level of the processing, ai(mod N) for each i is calculated to produce a number P of remainders; wherein at least one intermediate level of the processing is performed to reduce the P remainders to a single intermediate word, where each intermediate level of processing includes forming word groups from the remainder of a previous level and performing modulo N operations on the word groups; and wherein in a last level of the processing, a modulo N of the single intermediate word is calculated: (Wolrich, Abstract; FIG. 2; FIG. 3; FIG. 4; FIG. 6; FIG. 7; and paragraphs 0017-0043, where a process for determining a memory address for data in an interleaved memory system that does not use a power-of-two number of memory channels is described. During the process of Wolrich, an address is received as part of a request for access to data a memory location. The system of Wolrich first examines the received address to determine which memory channel should be used to route the request to memory and adjusts the received address to generate an intermediate address. The intermediate address then undergoes a justify operation to drop bits of the intermediate address to generate a justified address. The justified address is then processed to generate a final address at which the data may be accessed. Although the three-level processing of an interleaved memory address of Wolrich does not appear to explicitly teach the generation and use of modulo operations and remainders in order to calculate a modulo N value, it would have been obvious to a person having ordinary skill in the art to use the claimed multilevel address calculation operation. The modification would have been obvious because a person having ordinary skill in the art before the instant application was effectively filed would be motivated to increase system flexibility and performance by enabling usage of non-power-of-2 interleaving using computer processor hardware to perform binary operations to calculate addresses. Wolrich therefore particularly teaches wherein in a first level of the processing, ai(mod N) for each i is calculated to produce a number P of remainders; wherein at least one intermediate level of the processing is performed to reduce the P remainders to a single intermediate word, where each intermediate level of processing includes forming word groups from the remainder of a previous level and performing modulo N operations on the word groups; and wherein in a last level of the processing, a modulo N of the single intermediate word is calculated). It would have been obvious to a person having ordinary skill in the art, having the teachings of Wolrich, Avoinne, and Matloff before the instant application was effectively filed, to modify the combination of Avoinne with Matloff to include the principles of Wolrich of using multiple levels to calculate a memory address in a non-power-of-two interleaved memory system. The modification would have been obvious because a person having ordinary skill in the art would be motivated to increase system flexibility and performance by implementing techniques that enable memory interleaving in which any positive whole number of channels for interleaving memory may be used, thus enabling a higher level of concurrent access to the interleaved memory (Wolrich, paragraph 0010). None of Matloff, Avoinne, or Wolrich appears to teach the other limitations of this claim beyond those taught above; however, in an analogous art, Kalyanasundharam teaches nested channel address interleaving. As per claim 1, Kalyanasundharam particularly teaches: receiving a request to access the system memory, the request including an M-bit address that is decoded to determine a unique request to a cache line in the system memory: (Kalyanasundharam, Abstract; and paragraphs 0029-0030, where a memory access request for a specific memory address is received and looked up in cache to determine if the requested specific memory address is stored in cache. The Examiner notes that this means that the memory access request must be decoded in order to determine a memory address to access, and the decoded address is compared to memory addresses for data stored in cache to determine whether a cache hit occurs. Kalyanasundharam therefore particularly teaches receiving a request to access the system memory, the request including an M-bit address that is decoded to determine a unique request to a cache line in the system memory); processing the M-bit address to calculate a modulo N of the M-bit address to determine a remainder and use a lookup table (LUT), wherein the modulo N of the M-bit address is calculated and represents the M-bit address, wherein ai is an irh word in the address, 2bi is a power of two associated with the ith word, and module N of 2bi = 1 for all I and the remainder represents an offset, wherein in a last level of the processing, a modulo N of the single intermediate word is calculated and wherein a, b, i, and N are integers greater than or equal to zero: (Kalyanasundharam, Abstract; FIG. 5; FIG. 6; FIG. 7; and paragraphs 0043-0056, where the system of Kalyanasundharam interleaves memory addresses across a non-power-of-two number of channels using modulo division and a remainder as an offset into a table to determine in which partition and memory channel to locate a given interleaved memory address. Kalyanasundharam therefore particularly teaches processing the M-bit address to calculate a modulo N of the M-bit address to determine a remainder and use a lookup table (LUT), wherein the modulo N of the M-bit address is calculated and represents the M-bit address, wherein ai is an irh word in the address, 2bi is a power of two associated with the ith word, and module N of 2bi = 1 for all I and the remainder represents an offset, wherein in a last level of the processing, a modulo N of the single intermediate word is calculated and wherein a, b, i, and N are integers greater than or equal to zero). It would have been obvious to a person having ordinary skill in the art, having the teachings of Kalyanasundharam, Wolrich, Avoinne, and Matloff before them before the instant application was filed, to modify the combination of Wolrich with the combination of Avoinne with Matloff to include the principles of Kalyanasundharam of mapping an address space to a non-power-of-two number of memory channels for translation and interleaving such that each memory channel has an equal amount of mapped address space. The modification would have been obvious because a person having ordinary skill in the art would be motivated to increase system performance by implementing techniques for nested interleaving schemes that increase memory bandwidth by evenly servicing access across available memory (Kalyanasundharam, paragraph 0008). As per claim 2, the rejection of claim 1 is incorporated, and Wolrich further particularly teaches: wherein N is a prime number greater than 2: (Wolrich, paragraph 0010, where the number of memory channels may be equal to three. Wolrich therefore particularly teaches wherein N is a prime number greater than 2). As per claim 3, the rejection of claim 1 is incorporated, and Kalyanasundharam further particularly teaches: wherein a first intermediate level of processing includes forming P/2 intermediate word groups from the P remainder from the first level; and performing a modulo N operation on each of the P/2 intermediate word groups to produce P/2 remainders: (Kalyanasundharam, Fig. 5; FIG. 6; and paragraphs 0043-0050, where the system of Kalyanasundharam partitions a memory address space into regions based on a number of memory channels in order to enable even allocation of portions of the memory address space to a non-power-of-two number of memory channels. In the system of Kalyanasundharam, when a number of memory channels is 3, a memory address space may be divided into a 1/3 portion for one channel and a 2/3 portion to be divided between the remaining two channels. It would have been obvious to a person having ordinary skill in the art that a memory address space (and concomitant intermediate memory addresses) could be divided into quarter portions instead of one third and two third portions. The modification would have been obvious because a person having ordinary skill in the art would be motivated to increase system flexibility by evenly dividing memory address spaces among available memory channels. Kalyanasundharam therefore particularly teaches wherein a first intermediate level of processing includes forming P/2 intermediate word groups from the P remainder from the first level; and performing a modulo N operation on each of the P/2 intermediate word groups to produce P/2 remainders). As per claim 4, the rejection of claim 3 is incorporated, and Kalyanasundharam further particularly teaches: wherein a second intermediate level of processing includes forming P/4 intermediate word groups from the P/2 remainders from the first intermediate level; and performing a modulo N operation on each of the P/4 intermediate word groups to produce P/4 remainders: (Kalyanasundharam, Fig. 5; FIG. 6; and paragraphs 0043-0050, where the system of Kalyanasundharam partitions a memory address space into regions based on a number of memory channels in order to enable even allocation of portions of the memory address space to a non-power-of-two number of memory channels. In the system of Kalyanasundharam, when a number of memory channels is 3, a memory address space may be divided into a 1/3 portion for one channel and a 2/3 portion to be divided between the remaining two channels. It would have been obvious to a person having ordinary skill in the art that a memory address space (and concomitant intermediate memory addresses) could be divided into quarter portions instead of one third and two third portions. The modification would have been obvious because a person having ordinary skill in the art would be motivated to increase system flexibility by evenly dividing memory address spaces among available memory channels. Kalyanasundharam therefore particularly teaches wherein a second intermediate level of processing includes forming P/4 intermediate word groups from the P/2 remainders from the first intermediate level; and performing a modulo N operation on each of the P/4 intermediate word groups to produce P/4 remainders). As per claim 5, the rejection of claim 4 is incorporated, and Kalyanasundharam further particularly teaches: wherein the single intermediate word is formed from the P/4 intermediate word groups: (Kalyanasundharam, Fig. 5; FIG. 6; and paragraphs 0043-0050, where the system of Kalyanasundharam partitions a memory address space into regions based on a number of memory channels in order to enable even allocation of portions of the memory address space to a non-power-of-two number of memory channels. In the system of Kalyanasundharam, when a number of memory channels is 3, a memory address space may be divided into a 1/3 portion for one channel and a 2/3 portion to be divided between the remaining two channels. It would have been obvious to a person having ordinary skill in the art that a memory address space (and concomitant intermediate memory addresses) could be divided into quarter portions instead of one third and two third portions. The modification would have been obvious because a person having ordinary skill in the art would be motivated to increase system flexibility by evenly dividing memory address spaces among available memory channels. Kalyanasundharam therefore particularly teaches wherein the single intermediate word is formed from the P/4 intermediate word groups). As per claim 6, the rejection of claim 3 is incorporated, and Kalyanasundharam further particularly teaches: wherein a plurality of look up tables are used to perform the modulo N operations: (Kalyanasundharam, Fig. 5; FIG. 6; and paragraphs 0043-0050, where the system of Kalyanasundharam partitions a memory address space into regions based on a number of memory channels in order to enable even allocation of portions of the memory address space to a non-power-of-two number of memory channels. In the system of Kalyanasundharam, when a number of memory channels is 3, a memory address space may be divided into a 1/3 portion for one channel and a 2/3 portion to be divided between the remaining two channels. It would have been obvious to a person having ordinary skill in the art that a memory address space (and concomitant intermediate memory addresses) could be divided into quarter portions instead of one third and two third portions. The modification would have been obvious because a person having ordinary skill in the art would be motivated to increase system flexibility by evenly dividing memory address spaces among available memory channels. Kalyanasundharam therefore particularly teaches wherein a plurality of look up tables are used to perform the modulo N operations). As per claim 7, the rejection of claim 6 is incorporated, and Kalyanasundharam further particularly teaches: wherein each lookup table is a 4-bit table and each remainder is a 2-bit remainder: (Kalyanasundharam, Fig. 5; FIG. 6; and paragraphs 0043-0050, where the system of Kalyanasundharam partitions a memory address space into regions based on a number of memory channels in order to enable even allocation of portions of the memory address space to a non-power-of-two number of memory channels. In the system of Kalyanasundharam, when a number of memory channels is 3, a memory address space may be divided into a 1/3 portion for one channel and a 2/3 portion to be divided between the remaining two channels. It would have been obvious to a person having ordinary skill in the art that a memory address space (and concomitant intermediate memory addresses) could be divided into quarter portions instead of one third and two third portions. The modification would have been obvious because a person having ordinary skill in the art would be motivated to increase system flexibility by evenly dividing memory address spaces among available memory channels. Kalyanasundharam therefore particularly teaches wherein each lookup table is a 4-bit table and each remainder is a 2-bit remainder). As per claim 8, the rejection of claim 1 is incorporated, and Kalyanasundharam further particularly teaches: wherein the processing further includes using the modulo N of the single intermediate word to adjust the M-bit address to ensure that the M-bit address is divisible by N and also to determine which of the N interleaved channels to access: (Kalyanasundharam, Fig. 5; FIG. 6; and paragraphs 0043-0050, where the system of Kalyanasundharam partitions a memory address space into regions based on a number of memory channels in order to enable even allocation of portions of the memory address space to a non-power-of-two number of memory channels. In the system of Kalyanasundharam, when a number of memory channels is 3, a memory address space may be divided into a 1/3 portion for one channel and a 2/3 portion to be divided between the remaining two channels. It would have been obvious to a person having ordinary skill in the art that a memory address space (and concomitant intermediate memory addresses) could be divided into quarter portions instead of one third and two third portions. The modification would have been obvious because a person having ordinary skill in the art would be motivated to increase system flexibility by evenly dividing memory address spaces among available memory channels. Kalyanasundharam therefore particularly teaches wherein the processing further includes using the modulo N of the single intermediate word to adjust the M-bit address to ensure that the M-bit address is divisible by N and also to determine which of the N interleaved channels to access). As per claim 9, the rejection of claim 1 is incorporated, and Kalyanasundharam further particularly teaches: wherein the electronic system includes a network-on-chip having a plurality of network interface units; and wherein at least one of the network interface units is used to process the M-bit address to calculate the modulo N of the M-bit address: (Kalyanasundharam, Fig. 5; FIG. 6; and paragraphs 0043-0050, where the system of Kalyanasundharam partitions a memory address space into regions based on a number of memory channels in order to enable even allocation of portions of the memory address space to a non-power-of-two number of memory channels. In the system of Kalyanasundharam, when a number of memory channels is 3, a memory address space may be divided into a 1/3 portion for one channel and a 2/3 portion to be divided between the remaining two channels. It would have been obvious to a person having ordinary skill in the art that a memory address space (and concomitant intermediate memory addresses) could be divided into quarter portions instead of one third and two third portions. The modification would have been obvious because a person having ordinary skill in the art would be motivated to increase system flexibility by evenly dividing memory address spaces among available memory channels. Kalyanasundharam therefore particularly teaches wherein the electronic system includes a network-on-chip having a plurality of network interface units; and wherein at least one of the network interface units is used to process the M-bit address to calculate the modulo N of the M-bit address). As per claim 10, Matloff substantially teaches a network-on-chip (NoC) comprising: a transport interconnect; a target network interface unit (NIU) configured for interleaved N-channels system memory, where N is not a power of two; and a plurality of initiator NIUs for communicating with the target NIU via the transport interconnect: (Matloff, page 1, section 1 “Introduction,” paragraphs 1-3; and page 5, section “3.2 Prime Number of Modules,” where the computer system of Matloff comprises an interleaved memory system that utilizes a prime number A of memory modules (i.e., memory channels) that is not a power of 2. In addition, (Matloff, page 1, section 1 “Introduction,” paragraphs 1-3; and page 5, section “3.2 Prime Number of Modules”) teaches where the computer system of Matloff responds to a request for requested words of memory and thus must have received a memory request. The Examiner notes that the example memory architecture of Matloff uses power-of-two addressing which means that a memory address will comprise M bits. The Examiner further notes that receipt of a memory request requires both at least one sender of the memory request (i.e., an initiator NIU) and a target of the memory request (i.e., a target NIU). Finally, the Examiner notes that the system of Matloff must also comprise a transport network to enable communication and transfer of data between the sender of the memory request and the target of the memory request. Matloff therefore substantially teaches a transport interconnect; a target network interface unit (NIU) configured for interleaved N-channels system memory, where N is not a power of two; and a plurality of initiator NIUs for communicating with the target NIU via the transport interconnect). Matloff does not appear to explicitly teach the other limitations of this claim beyond those taught above; however, in an analogous art, Avoinne teaches memory interleaving on memory channels. As per claim 10, Avoinne particularly teaches: wherein at least one of the NIUs is configured to process an address to calculate a modulo N of an M-bit address without performing division-by-N, wherein the modulo N of the address is calculated as [∑iai mod N * 2bi mod N] mod N, where ∑ai * 2bi represents the M-bit address, ai is an ith word in the address, 2bi is a power of two associated with the ith word, and mod N of 2bi = 1 for all i: (Avoinne, Abstract; FIG. 5; and paragraphs 0006 and 0026-0028, where the system of Avoinne merely examines received bits of an address in order to determine a channel to which the address should be routed. Although Avoinne does not appear to explicitly teach using the claimed equation of [∑iaimod N * 2bi mod N] mod N, where ∑iai * 2bi represents the M-bit address, ai is an ith word in the address, 2bi is a power of two associated with the ith word, and modulo N of 2bi = 1 for all i in order to calculate the modulo N of the address, it would have been obvious to a person having ordinary skill in the art to use the claimed equation to calculate the modulo N of the address. The modification would have been obvious because a person having ordinary skill in the art before the instant application was effectively filed would be motivated to increase system flexibility and performance by enabling usage of non-power-of-2 interleaving using computer processor hardware to perform binary operations to calculate addresses. Avoinne therefore particularly teaches wherein at least one of the NIUs is configured to process an address to calculate a modulo N of an M-bit address without performing division-by-N, wherein the modulo N of the address is calculated as [∑iai mod N * 2bi mod N] mod N, where ∑ai * 2bi represents the M-bit address, ai is an ith word in the address, 2bi is a power of two associated with the ith word, and mod N of 2bi = 1 for all i). It would have been obvious to a person having ordinary skill in the art, having the teachings of Avoinne and Matloff before them before the instant application was effectively filed, to modify the system of Matloff to use the principles of Avoinne of hardware memory interleaving. The modification would have been obvious because a person having ordinary skill in the art would be motivated to increase system performance by implementing techniques to perform memory interleaving in hardware. Neither Matloff nor Avoinne appears to explicitly teach the other limitations of this claim beyond those taught above; however, in an analogous art, Wolrich teaches memory interleaving. As per claim 10, Wolrich particularly teaches: wherein in a first level of the processing, ai(mod N) for each i is calculated to produce a number P of remainders; wherein at least one intermediate level of the processing is performed to reduce the P remainders to a single intermediate word, wherein each intermediate level of the processing includes forming word groups from the remainders of a previous level and performing modulo N operations on the word groups; and wherein in a last level of the processing, a modulo N of the single intermediate word is calculated: (Wolrich, Abstract; FIG. 2; FIG. 3; FIG. 4; FIG. 6; FIG. 7; and paragraphs 0017-0043, where a process for determining a memory address for data in an interleaved memory system that does not use a power-of-two number of memory channels is described. During the process of Wolrich, an address is received as part of a request for access to data a memory location. The system of Wolrich first examines the received address to determine which memory channel should be used to route the request to memory and adjusts the received address to generate an intermediate address. The intermediate address then undergoes a justify operation to drop bits of the intermediate address to generate a justified address. The justified address is then processed to generate a final address at which the data may be accessed. Although the three-level processing of an interleaved memory address of Wolrich does not appear to explicitly teach the generation and use of modulo operations and remainders in order to calculate a modulo N value, it would have been obvious to a person having ordinary skill in the art to use the claimed multilevel address calculation operation. The modification would have been obvious because a person having ordinary skill in the art before the instant application was effectively filed would be motivated to increase system flexibility and performance by enabling usage of non-power-of-2 interleaving using computer processor hardware to perform binary operations to calculate addresses. Wolrich therefore particularly teaches wherein in a first level of the processing, ai(mod N) for each i is calculated to produce a number P of remainders; wherein at least one intermediate level of the processing is performed to reduce the P remainders to a single intermediate word, wherein each intermediate level of the processing includes forming word groups from the remainders of a previous level and performing modulo N operations on the word groups; and wherein in a last level of the processing, a modulo N of the single intermediate word is calculated). It would have been obvious to a person having ordinary skill in the art, having the teachings of Wolrich, Avoinne, and Matloff before the instant application was effectively filed, to modify the combination of Avoinne with Matloff to include the principles of Wolrich of using multiple levels to calculate a memory address in a non-power-of-two interleaved memory system. The modification would have been obvious because a person having ordinary skill in the art would be motivated to increase system flexibility and performance by implementing techniques that enable memory interleaving in which any positive whole number of channels for interleaving memory may be used, thus enabling a higher level of concurrent access to the interleaved memory (Wolrich, paragraph 0010). None of Matloff, Avoinne, or Wolrich appears to teach the other limitations of this claim beyond those taught above; however, in an analogous art, Kalyanasundharam teaches nested channel address interleaving. As per claim 10, Kalyanasundharam particularly teaches: receiving a request to access the system memory, the request including an M-bit address that is decoded to determine a unique request to a cache line in the system memory: (Kalyanasundharam, Abstract; and paragraphs 0029-0030, where a memory access request for a specific memory address is received and looked up in cache to determine if the requested specific memory address is stored in cache. The Examiner notes that this means that the memory access request must be decoded in order to determine a memory address to access, and the decoded address is compared to memory addresses for data stored in cache to determine whether a cache hit occurs. Kalyanasundharam therefore particularly teaches receiving a request to access the system memory, the request including an M-bit address that is decoded to determine a unique request to a cache line in the system memory); processing the M-bit address to calculate a modulo N of the M-bit address to determine a remainder and use a lookup table (LUT), wherein the modulo N of the M-bit address is calculated and represents the M-bit address, wherein ai is an irh word in the address, 2bi is a power of two associated with the ith word, and module N of 2bi = 1 for all I and the remainder represents an offset, wherein in a last level of the processing, a modulo N of the single intermediate word is calculated and wherein a, b, i, and N are integers greater than or equal to zero: (Kalyanasundharam, Abstract; FIG. 5; FIG. 6; FIG. 7; and paragraphs 0043-0056, where the system of Kalyanasundharam interleaves memory addresses across a non-power-of-two number of channels using modulo division and a remainder as an offset into a table to determine in which partition and memory channel to locate a given interleaved memory address. Kalyanasundharam therefore particularly teaches processing the M-bit address to calculate a modulo N of the M-bit address to determine a remainder and use a lookup table (LUT), wherein the modulo N of the M-bit address is calculated and represents the M-bit address, wherein ai is an irh word in the address, 2bi is a power of two associated with the ith word, and module N of 2bi = 1 for all I and the remainder represents an offset, wherein in a last level of the processing, a modulo N of the single intermediate word is calculated and wherein a, b, i, and N are integers greater than or equal to zero). It would have been obvious to a person having ordinary skill in the art, having the teachings of Kalyanasundharam, Wolrich, Avoinne, and Matloff before them before the instant application was filed, to modify the combination of Wolrich with the combination of Avoinne with Matloff to include the principles of Kalyanasundharam of mapping an address space to a non-power-of-two number of memory channels for translation and interleaving such that each memory channel has an equal amount of mapped address space. The modification would have been obvious because a person having ordinary skill in the art would be motivated to increase system performance by implementing techniques for nested interleaving schemes that increase memory bandwidth by evenly servicing access across available memory (Kalyanasundharam, paragraph 0008). As per claim 11, the rejection of claim 10 is incorporated, and Wolrich further particularly teaches: wherein N is a prime number greater than 2: (Wolrich, paragraph 0010, where the number of memory channels may be equal to three. Wolrich therefore particularly teaches wherein N is a prime number greater than 2). As per claim 12, the rejection of claim 10 is incorporated, and the Examiner notes that the language of dependent claim 12 is substantially similar to the language of dependent claim 3. Dependent claim 12 is therefore rejected using the same references and reasoning, mutatis mutandis, as used in the above rejection of dependent claim 3. As per claim 13, the rejection of claim 12 is incorporated, and the Examiner notes that the language of dependent claim 13 is substantially similar to the language of dependent claim 4. Dependent claim 13 is therefore rejected using the same references and reasoning, mutatis mutandis, as used in the above rejection of dependent claim 4. As per claim 14, the rejection of claim 13 is incorporated, and the Examiner notes that the language of dependent claim 14 is substantially similar to the language of dependent claim 5. Dependent claim 14 is therefore rejected using the same references and reasoning, mutatis mutandis, as used in the above rejection of dependent claim 5. As per claim 15, the rejection of claim 12 is incorporated, and the Examiner notes that the language of dependent claim 15 is substantially similar to the language of dependent claim 6. Dependent claim 15 is therefore rejected using the same references and reasoning, mutatis mutandis, as used in the above rejection of dependent claim 6. As per claim 16, the rejection of claim 15 is incorporated, and the Examiner notes that the language of dependent claim 16 is substantially similar to the language of dependent claim 7. Dependent claim 16 is therefore rejected using the same references and reasoning, mutatis mutandis, as used in the above rejection of dependent claim 7. As per claim 17, the rejection of claim 10 is incorporated, and the Examiner notes that the language of dependent claim 17 is substantially similar to the language of dependent claim 8. Dependent claim 17 is therefore rejected using the same references and reasoning, mutatis mutandis, as used in the above rejection of dependent claim 8. As per claim 18, the rejection of claim 10 is incorporated, and Kalyanasundharam further particularly teaches: wherein at least one of the NIUs includes an initiator network interface unit configured to process the M-bit address to calculate the module N of the M-bit address: (Kalyanasundharam, Fig. 5; FIG. 6; and paragraphs 0043-0050, where the system of Kalyanasundharam partitions a memory address space into regions based on a number of memory channels in order to enable even allocation of portions of the memory address space to a non-power-of-two number of memory channels. In the system of Kalyanasundharam, when a number of memory channels is 3, a memory address space may be divided into a 1/3 portion for one channel and a 2/3 portion to be divided between the remaining two channels. It would have been obvious to a person having ordinary skill in the art that a memory address space (and concomitant intermediate memory addresses) could be divided into quarter portions instead of one third and two third portions. The modification would have been obvious because a person having ordinary skill in the art would be motivated to increase system flexibility by evenly dividing memory address spaces among available memory channels. Kalyanasundharam therefore particularly teaches wherein at least one of the NIUs includes an initiator network interface unit configured to process the M-bit address to calculate the module N of the M-bit address). As per claim 19, the rejection of claim 10 is incorporated, and Kalyanasundharam further particularly teaches: wherein at least one of the NIUs includes a target network interface unit configured to process the M-bit <address?> to calculate the module N of the M-bit address: (Kalyanasundharam, Fig. 5; FIG. 6; and paragraphs 0043-0050, where the system of Kalyanasundharam partitions a memory address space into regions based on a number of memory channels in order to enable even allocation of portions of the memory address space to a non-power-of-two number of memory channels. In the system of Kalyanasundharam, when a number of memory channels is 3, a memory address space may be divided into a 1/3 portion for one channel and a 2/3 portion to be divided between the remaining two channels. It would have been obvious to a person having ordinary skill in the art that a memory address space (and concomitant intermediate memory addresses) could be divided into quarter portions instead of one third and two third portions. The modification would have been obvious because a person having ordinary skill in the art would be motivated to increase system flexibility by evenly dividing memory address spaces among available memory channels. Kalyanasundharam therefore particularly teaches wherein at least one of the NIUs includes a target network interface unit configured to process the M-bit <address?> to calculate the module N of the M-bit address). As per claim 20, Matloff substantially teaches a system on chip (SoC) comprising: system memory having N interleaved channels, where N is not a power of two; and a network-on-chip having a plurality of network interface units; wherein at least one of the network interface units is configured to: (Matloff, page 1, section 1 “Introduction,” paragraphs 1-3; and page 5, section “3.2 Prime Number of Modules,” where the computer system of Matloff comprises an interleaved memory system that utilizes a prime number A of memory modules (i.e., memory channels) that is not a power of 2. In addition, (Matloff, page 1, section 1 “Introduction,” paragraphs 1-3; and page 5, section “3.2 Prime Number of Modules”) teaches where the computer system of Matloff responds to a request for requested words of memory and thus must have received a memory request. The Examiner notes that the example memory architecture of Matloff uses power-of-two addressing which means that a memory address will comprise M bits. The Examiner further notes that receipt of a memory request requires both at least one sender of the memory request (i.e., an initiator network interface unit) and a target of the memory request (i.e., a target network interface unit). Finally, the Examiner notes that the system of Matloff must also comprise a transport network to enable communication and transfer of data between the sender of the memory request and the target of the memory request. Matloff therefore substantially teaches system memory having N interleaved channels, where N is not a power of two; and a network-on-chip having a plurality of network interface units; wherein at least one of the network interface units is configured to); receive a request to access the system memory, the request including an M-bit address: (Matloff, page 1, section 1 “Introduction,” paragraphs 1-3; and page 5, section “3.2 Prime Number of Modules,” where the computer system of Matloff responds to a request for requested words of memory and thus must have received a memory request. The Examiner notes that the example memory architecture of Matloff uses power-of-two addressing which means that a memory address will comprise M bits. Matloff therefore substantially teaches receive a request to access the system memory, the request including an M-bit address). Matloff does not appear to explicitly teach the other limitations of this claim beyond those taught above; however, in an analogous art, Avoinne teaches memory interleaving on memory channels. As per claim 20, Avoinne particularly teaches: and process the address to calculate a modulo N of the address without performing division-by-N, wherein the modulo N of the address is calculated as [∑iai mod N * 2b * mod N] mod N, where ∑iai * 2bi represents the M-bit address, ai is an ith word in the address, 2bi is a power of two associated with the ith word, and mod N of 2bi = 1 for all i: (Avoinne, Abstract; FIG. 5; and paragraphs 0006 and 0026-0028, where the system of Avoinne merely examines received bits of an address in order to determine a channel to which the address should be routed. Although Avoinne does not appear to explicitly teach using the claimed equation of [∑iaimod N * 2bi mod N] mod N, where ∑iai * 2bi represents the M-bit address, ai is an ith word in the address, 2bi is a power of two associated with the ith word, and modulo N of 2bi = 1 for all i in order to calculate the modulo N of the address, it would have been obvious to a person having ordinary skill in the art to use the claimed equation to calculate the modulo N of the address. The modification would have been obvious because a person having ordinary skill in the art before the instant application was effectively filed would be motivated to increase system flexibility and performance by enabling usage of non-power-of-2 interleaving using computer processor hardware to perform binary operations to calculate addresses. Avoinne therefore particularly teaches and process the address to calculate a modulo N of the address without performing division-by-N, wherein the modulo N of the address is calculated as [∑iai mod N * 2b * mod N] mod N, where ∑iai * 2bi represents the M-bit address, ai is an ith word in the address, 2bi is a power of two associated with the ith word, and mod N of 2bi = 1 for all i). It would have been obvious to a person having ordinary skill in the art, having the teachings of Avoinne and Matloff before them before the instant application was effectively filed, to modify the system of Matloff to use the principles of Avoinne of hardware memory interleaving. The modification would have been obvious because a person having ordinary skill in the art would be motivated to increase system performance by implementing techniques to perform memory interleaving in hardware. Neither Matloff nor Avoinne appears to explicitly teach the other limitations of this claim beyond those taught above; however, in an analogous art, Wolrich teaches memory interleaving. As per claim 20, Wolrich particularly teaches: wherein: in a first level of processing, ai(mod N) is calculated; in at least one intermediate level of processing, remainders from a previous level are formed into intermediate word groups, and modulo N operations are performed on the intermediate word groups until a single intermediate word is formed; and in a last level of processing, a modulo N of the single intermediate word is performed: (Wolrich, Abstract; FIG. 2; FIG. 3; FIG. 4; FIG. 6; FIG. 7; and paragraphs 0017-0043, where a process for determining a memory address for data in an interleaved memory system that does not use a power-of-two number of memory channels is described. During the process of Wolrich, an address is received as part of a request for access to data a memory location. The system of Wolrich first examines the received address to determine which memory channel should be used to route the request to memory and adjusts the received address to generate an intermediate address. The intermediate address then undergoes a justify operation to drop bits of the intermediate address to generate a justified address. The justified address is then processed to generate a final address at which the data may be accessed. Although the three-level processing of an interleaved memory address of Wolrich does not appear to explicitly teach the generation and use of modulo operations and remainders in order to calculate a modulo N value, it would have been obvious to a person having ordinary skill in the art to use the claimed multilevel address calculation operation. The modification would have been obvious because a person having ordinary skill in the art before the instant application was effectively filed would be motivated to increase system flexibility and performance by enabling usage of non-power-of-2 interleaving using computer processor hardware to perform binary operations to calculate addresses. Wolrich therefore particularly teaches wherein: in a first level of processing, ai(mod N) is calculated; in at least one intermediate level of processing, remainders from a previous level are formed into intermediate word groups, and modulo N operations are performed on the intermediate word groups until a single intermediate word is formed; and in a last level of processing, a modulo N of the single intermediate word is performed). It would have been obvious to a person having ordinary skill in the art, having the teachings of Wolrich, Avoinne, and Matloff before the instant application was effectively filed, to modify the combination of Avoinne with Matloff to include the principles of Wolrich of using multiple levels to calculate a memory address in a non-power-of-two interleaved memory system. The modification would have been obvious because a person having ordinary skill in the art would be motivated to increase system flexibility and performance by implementing techniques that enable memory interleaving in which any positive whole number of channels for interleaving memory may be used, thus enabling a higher level of concurrent access to the interleaved memory (Wolrich, paragraph 0010). None of Matloff, Avoinne, or Wolrich appears to teach the other limitations of this claim beyond those taught above; however, in an analogous art, Kalyanasundharam teaches nested channel address interleaving. As per claim 20, Kalyanasundharam particularly teaches: receiving a request to access the system memory, the request including an M-bit address that is decoded to determine a unique request to a cache line in the system memory: (Kalyanasundharam, Abstract; and paragraphs 0029-0030, where a memory access request for a specific memory address is received and looked up in cache to determine if the requested specific memory address is stored in cache. The Examiner notes that this means that the memory access request must be decoded in order to determine a memory address to access, and the decoded address is compared to memory addresses for data stored in cache to determine whether a cache hit occurs. Kalyanasundharam therefore particularly teaches receiving a request to access the system memory, the request including an M-bit address that is decoded to determine a unique request to a cache line in the system memory); processing the M-bit address to calculate a modulo N of the M-bit address to determine a remainder and use a lookup table (LUT), wherein the modulo N of the M-bit address is calculated and represents the M-bit address, wherein ai is an irh word in the address, 2bi is a power of two associated with the ith word, and module N of 2bi = 1 for all I and the remainder represents an offset, wherein in a last level of the processing, a modulo N of the single intermediate word is calculated and wherein a, b, i, and N are integers greater than or equal to zero: (Kalyanasundharam, Abstract; FIG. 5; FIG. 6; FIG. 7; and paragraphs 0043-0056, where the system of Kalyanasundharam interleaves memory addresses across a non-power-of-two number of channels using modulo division and a remainder as an offset into a table to determine in which partition and memory channel to locate a given interleaved memory address. Kalyanasundharam therefore particularly teaches processing the M-bit address to calculate a modulo N of the M-bit address to determine a remainder and use a lookup table (LUT), wherein the modulo N of the M-bit address is calculated and represents the M-bit address, wherein ai is an irh word in the address, 2bi is a power of two associated with the ith word, and module N of 2bi = 1 for all I and the remainder represents an offset, wherein in a last level of the processing, a modulo N of the single intermediate word is calculated and wherein a, b, i, and N are integers greater than or equal to zero). It would have been obvious to a person having ordinary skill in the art, having the teachings of Kalyanasundharam, Wolrich, Avoinne, and Matloff before them before the instant application was filed, to modify the combination of Wolrich with the combination of Avoinne with Matloff to include the principles of Kalyanasundharam of mapping an address space to a non-power-of-two number of memory channels for translation and interleaving such that each memory channel has an equal amount of mapped address space. The modification would have been obvious because a person having ordinary skill in the art would be motivated to increase system performance by implementing techniques for nested interleaving schemes that increase memory bandwidth by evenly servicing access across available memory (Kalyanasundharam, paragraph 0008). Response to Arguments In the Remarks dated 6/24/2026, Applicant substantially argues: None of the prior art, alone or in combination, teaches or suggests the claimed combination of features of the as-amended claims of the instant application. Applicant’s arguments dated 6/24/2026 have been fully considered, but they are moot in view of the new grounds of rejection that were necessitated by Applicant’s amendments to the claims. The Examiner notes that the new Kalyanasundharam reference in combination with the other prior art of record clearly teaches the features added via Applicant’s amendments to the claims. The Examiner further notes that the new grounds of rejection were necessitated by Applicant’s amendments to the claims. 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 Daniel C. Chappell whose telephone number is (571)272-5003. The examiner can normally be reached 1000-1800, Eastern. 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, Jared I. Rutz can be reached at (571)272-5535. 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. Daniel C. Chappell Primary Examiner Art Unit 2135 /Daniel C. Chappell/Primary Examiner, Art Unit 2135
Read full office action

Prosecution Timeline

Feb 03, 2025
Application Filed
Feb 24, 2026
Non-Final Rejection mailed — §103, §112
Mar 20, 2026
Interview Requested
Apr 22, 2026
Examiner Interview Summary
Apr 22, 2026
Applicant Interview (Telephonic)
Jun 24, 2026
Response Filed
Aug 26, 2026
Final Rejection mailed — §103, §112 (current)

Precedent Cases

Applications granted by this same examiner with similar technology

Patent 12792240
ADAPTIVE CONTROL FOR IN-MEMORY VERSIONING
1y 9m to grant Granted Nov 10, 2026
Patent 12783404
COMMAND-DIFFERENTIATED STORAGE OF INTERNALLY AND EXTERNALLY SOURCED DATA
1y 11m to grant Granted Nov 03, 2026
Patent 12762011
CONFIGURABLE CACHE FOR COHERENT SYSTEM
1y 11m to grant Granted Oct 13, 2026
Patent 12730580
SYSTEMS AND METHODS FOR MANAGING STORAGE SYSTEM MONITORING DATA USING PREDICTIVE COMPRESSION
1y 11m to grant Granted Sep 08, 2026
Patent 12730931
GENERATION OF VECTORS FOR RETRIEVAL AUGMENTED GENERATION USING BACKUP DATA
1y 7m to grant Granted Sep 08, 2026
Study what changed to get past this examiner. Based on 5 most recent grants.

Strategy Recommendation AI-generated — please review before filing

Get a prosecution strategy drawn from examiner precedents, rejection analysis, and claim mapping.
Typically takes 5-10 seconds — AI-generated, attorney review required before filing

Prosecution Projections

3-4
Expected OA Rounds
81%
Grant Probability
99%
With Interview (+45.1%)
Moderate
PTA Risk
Based on 618 resolved cases by this examiner. Grant probability derived from career allowance rate.

Sign in with your work email

Enter your email to receive a magic link. No password needed.

Personal email addresses (Gmail, Yahoo, etc.) are not accepted.

Free tier: 3 strategy analyses per month