Jin-Ki Kim - Ottawa, CA HakJune Oh - Kanata, CA Hong Beom Pyeon - Kanata, CA Steven Przybylski - Ann Arbor MI, US
Assignee:
Mosaid Technologies Incorporated - Ottawa
International Classification:
G06F 12/00
US Classification:
711100, 711 5, 711103, 711154
Abstract:
A memory system architecture has serially connected memory devices. The memory system is scalable to include any number of memory devices without any performance degradation or complex redesign. Each memory device has a serial input/output interface for communicating between other memory devices and a memory controller. The memory controller issues commands in at least one bitstream, where the bitstream follows a modular command protocol. The command includes an operation code with optional address information and a device address, so that only the addressed memory device acts upon the command. Separate data output strobe and command input strobe signals are provided in parallel with each output data stream and input command data stream, respectively, for identifying the type of data and the length of the data. The modular command protocol is used for executing concurrent operations in each memory device to further improve performance.
Error Detection And Correction Codes For Channels And Memories With Incomplete Error Characteristics
A channel has a first and a second end. The first end of the channel is coupled to a transmitter. The channel is capable of transmitting symbols selected from a symbol set from the first end to the second end. The channel exhibits incomplete error introduction properties. A code comprises a set of code words. The elements of the set of code words are one or more code symbols long. The code symbols are members of the symbol set. The minimum modified Hamming separation between the elements of the set of code words in light of the error introduction properties of the channel is greater than the minimum Hamming distance between the elements of the set of code words. A memory device, a method of using the channel, and a method of generating the code are also described.
In a ring-of-clusters network topology, groups of slave devices are accessed in parallel, such that the latency around the ring is proportional to the number of clusters and not proportional to the number of integrated circuits. The devices of a cluster share input and output ring segments such that packets arriving on the input segment are received and interpreted by all the devices in a cluster. In other embodiments, none, some or all but one slaves per cluster are asleep or otherwise disabled so that they do not input and interpret incoming packets. Regardless, in all embodiments, the slaves of a cluster cooperate, potentially under the controller's direction, to ensure that at most one of them is actively driving the output segment at any given time. The devices may be addressed through a device ID, a cluster ID, or a combination thereof. Embodiments of the invention are suited to exploit multi-chip module implementations and forms of vertical circuit stacking.
Id Generation Apparatus And Method For Serially Interconnected Devices
Hong Beom PYEON - Kanata, CA HakJune OH - Kanata, CA Jin-Ki KIM - Ottawa, CA Steven A. PRZYBYLSKI - Ann Arbor MI, US
Assignee:
MOSAID TECHNOLOGIES INCORPORATED - Kanata ON
International Classification:
G11C 8/04
US Classification:
711170
Abstract:
A plurality of memory devices (e.g., DRAMs, SRAMs, NAND Flash, NOR Flash) is serially interconnected. Each of the interconnected devices receives a device identifier (ID) and latches it as its ID. Each device includes a circuit for calculating another ID or an incremented ID to generate it. The generated ID is transferred to another device and the ID is incremented in each of the devices in the serial interconnection. The last device in the interconnection provides a last generated ID that is provided to a memory controller having a recognition circuit that recognizes the total number of the serially interconnected devices, from the provided last generated ID. The recognition circuit recognizes the total output latency of the devices in the serial interconnection.
Steven PRZYBYLSKI - Ann Arbor MI, US Roland SCHUETZ - Ottawa, CA HakJune OH - Kanata, CA Hong Beom PYEON - Kanata, CA
Assignee:
MOSAID TECHNOLOGIES INCORPORATED - Kanata
International Classification:
G06F 12/00
US Classification:
711171, 711E12001
Abstract:
A system having serially connected memory devices in a ring topology organization to realize high speed performance. The memory devices have dynamically configurable data widths such that the system can operate with up to a maximum common number of active data pads to maximize performance, or to operate with a single active data pad to minimize power consumption. Therefore the system can include a mix of memory devices having different data widths. The memory devices are dynamically configurable through the issuance of a single command propagated serially through all the memory devices from the memory controller in a broadcast operation. Robust operation of the system is ensured by implementing a data output inhibit algorithm, which prevents valid data from being provided to the memory controller when read output control signal is received out of its proper sequence.
MOSAID TECHNOLOGIES INCORPORATED - Kanata, CA HakJune OH - Kanata, CA Hong Beom PYEON - Kanata, CA Steven PRZYBYLSKI - Ann Arbor MI, US
Assignee:
MOSAID TECHNOLOGIES INCORPORATED - Kanata
International Classification:
G11C 16/10
US Classification:
36518511
Abstract:
A memory system architecture has serially connected memory devices. The memory system is scalable to include any number of memory devices without any performance degradation or complex redesign. Each memory device has a serial input/output interface for communicating between other memory devices and a memory controller. The memory controller issues commands in at least one bitstream, where the bitstream follows a modular command protocol. The command includes an operation code with optional address information and a device address, so that only the addressed memory device acts upon the command. Separate data output strobe and command input strobe signals are provided in parallel with each output data stream and input command data stream, respectively, for identifying the type of data and the length of the data. The modular command protocol is used for executing concurrent operations in each memory device to further improve performance.
Error Detection And Correction Codes For Channels And Memories With Incomplete Error Characteristics
- Ottawa ON, CA Steven PRZYBYLSKI - Ann Arbor MI, US
Assignee:
MOSAID TECHNOLOGIES INCORPORATED - Ottawa
International Classification:
G11C 29/00
US Classification:
714773, 714763
Abstract:
A channel has a first and a second end. The first end of the channel is coupled to a transmitter. The channel is capable of transmitting symbols selected from a symbol set from the first end to the second end. The channel exhibits incomplete error introduction properties. A code comprises a set of code words. The elements of the set of code words are one or more code symbols long. The code symbols are members of the symbol set. The minimum modified Hamming separation between the elements of the set of code words in light of the error introduction properties of the channel is greater than the minimum Hamming distance between the elements of the set of code words. A memory device, a method of using the channel, and a method of generating the code are also described.
Cup Chip Having Tag Comparator And Address Translation Unit On Chip And Connected To Off-Chip Cache And Main Memories
John P. Moussouris - Palo Alto CA Lester M. Crudele - San Jose CA Steven A. Przybylski - Menlo Park CA
Assignee:
MIPS Computer Systems, Inc. - Mountain View CA
International Classification:
G06F 934 G06F 1210 G06F 1316
US Classification:
364200
Abstract:
A cache-based computer architecture has the address generating unit and the tag comparator packaged together and separately from the cache RAMS. If the architecture supports virtual memory, an address translation unit may be included on the same chip as, and logically between, the address generating unit and the tag comparator logic. Further, interleaved access to more than one cache may be accomplished on the external address, data and tag busses.
University of California, Berkeley, Haas School of Business 2000 - 2000
Master of Business Administration, Masters
Stanford University 1988 - 1988
Doctorates, Doctor of Philosophy, Electrical Engineering
Stanford University 1982 - 1982
Masters, Master of Science In Electrical Engineering
University of Toronto 1980 - 1980
Bachelors, Bachelor of Arts, Computer Science, Engineering
Duro-Last Roofing, Inc Saginaw, MI May 2009 to Sep 2011 National Manufacturing Operations ManagerDelphi Corporation, Saginaw Steering Systems Saginaw, MI Apr 2002 to May 2009 Plant Quality Assurance ManagerDelphi Corporation, Saginaw Steering Systems Saginaw, MI Dec 2000 to Apr 2002 Plant Problem Solving, Lean Manufacturing, & Continuous Improvement Leader (Six Sigma Blackbelt)Delphi Corporation, Saginaw Steering Systems Saginaw, MI Mar 2000 to Dec 2000 Plant Skilled Trades/Maintenance ManagerDelphi Corporation, Saginaw Steering Systems Saginaw, MI Oct 1998 to Mar 2000 Plant Manufacturing Operations ManagerDelphi Corporation, Saginaw Steering Systems Saginaw, MI Aug 1995 to Oct 1998 Senior Manufacturing EngineerDelphi Corporation, Saginaw Steering Systems Saginaw, MI Jan 1992 to Aug 1995 Manufacturing Engineer
Education:
University of Michigan Ann Arbor, MI Dec 1991 B.S. in Mechanical Engineering
Skills:
Six Sigma Blackbelt, Shainin Red X Master, Lean Manufacturing & Process Continuous Improvement