• KIISE Transactions on Computing Practices
• /
• v.23 no.8
• /
• pp.466-472
• /
• 2017

Hardware Transactional Memory (HTM) has greatly changed the parallel programming paradigm for transaction processing. Since Intel has recently proposed Transactional Synchronization Extension (TSX), a number of studies based on HTM have been conducted. However, the existing studies support conflict prediction for a single cause of the transaction processing and provide a standardized TSX environment for all workloads. To solve the problems, we propose an efficient hardware transactional memory scheme for processing transactions in multi-core in-memory environment. First, the proposed scheme determines whether to use Software Transactional Memory (STM) or the serial execution as a fallback path of HTM by using a prediction matrix to collect the information of previously executed transactions. Second, the proposed scheme performs efficient transaction processing according to the characteristic of a given workload by providing a retry policy based on machine learning algorithms. Finally, through the experimental performance evaluation using Stanford transactional applications for multi-processing (STAMP), the proposed scheme shows 10~20% better performance than the existing schemes.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.45 no.4
• /
• pp.35-41
• /
• 2008

When memory devices are exposed to a space environment. they suffer various effects such as SEU(Single Event Upset). For these reasons, memory systems for space applications are generally equipped with error detection and correction(EDAC) logics against SEUs. In this paper, the error detection and correction strategy in the Mass Memory Unit(MMU) of the STSAT-3 is discussed. The probability equation of un-recoverable SEUs in the mass memory system is derived when the whole memory is encoded and decoded by the RS(10,8) Reed-Solomon code. Also the probability value is analyzed for various occurrence rates of SEUs which the STSAT-3 possibly suffers. The analyzed results can be used to determine the period of scrubbing the whole memory, which is one of the important parameters in the design of the MMU.

• The KIPS Transactions:PartA
• /
• v.8A no.4
• /
• pp.419-428
• /
• 2001

Advances in circuit and integration technology are continuously boosting the speed of processors. One of the main challenges presented by such developments is the effective use of powerful processors in shared memory multiprocessor system. We believe that the interconnection problem is not solved even for small scale shared memory multiprocessor, since the speed of shared buses is unlikely to keep up with the bandwidth requirements of new powerful processors. In the past few years, point-to-point unidirectional connection have emerged as a very promising interconnection technology. The single slotted ring is the simplest form point-to-point interconnection. The main limitation of the single slotted ring architecture is that latency of access increase linearly with the number of the processors in the ring. Because of this, we proposed the dual slotted ring as an alternative to single slotted ring for cache-based multiprocessor system. In this paper, we analyze the proposed dual slotted ring architecture using new snooping protocol and enforce simulation to compare it with single slotted ring.

• Journal of the Korean Vacuum Society
• /
• v.14 no.1
• /
• pp.40-48
• /
• 2005

Bucky shuttle memory systems were investigated by the classical molecular dynamics(MD) simulations. Energetics and operating response of the shuttle-memory-elements u?ere examined by MD simulations of the C/sub 60/ shuttle in the nanomemory systems under various external force fields. Single-nanopeapod type was consisting of three fullerenes encapsulated in (10, 10) boron-nitride nanotube and filled Cu electrode. Studied systems could be applied to nonvolatile memory. MD simulation results showed that the stable bit flops could be achieved from the external force fields of 0.1 eV/Å for single-nanopeapod type.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2007.06a
• /
• pp.20-20
• /
• 2007

Nanowire-based field-effect transistors (FETs) decorated with nanoparticles have been greatly paid attention as nonvolatile memory devices of next generation due to their excellent transportation ability of charge carriers in the channel and outstanding capability of charge trapping in the floating gate. In this work, top-gate single ZnO nanowire-based FETs with and without Au nanoparticles were fabricated and their memory effects were characterized. Using thermal evaporation and rapid thermal annealing processes, Au nanoparticles were formed on an $Al_2O_3$ layer which was semi cylindrically coated on a single ZnO nanowire. The family of $I_{DS}-V_{GS}$ curves for the double sweep of the gate voltage at $V_{DS}$ = 1 V was obtained. The device decorated with nanoparticles shows giant hysterisis loops with ${\Delta}V_{th}$ = 2 V, indicating a significant charge storage effect. Note that the hysterisis loops are clockwise which result from the tunneling of the charge carriers from the nanowire into the nanoparticles. On the other hand, the device without nanoparticles shows a negligible countclockwise hysterisis loop which reveals that the influence of oxide trap charges or mobile ions is negligible. Therefore, the charge storage effect mainly comes from the nanoparticles decorated on the nanowire, which obviously demonstrates that the top-gate single ZnO nanowire-based FETs decorated with Au nanoparticles are the good candidate for the application in the nonvolatile memory devices of next generation.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.24 no.10 s.181
• /
• pp.2461-2468
• /
• 2000

In this paper, single fiber pull-out test is used to measure the interfacial bonding shear strength of $Ti_{50}-Ni_{50}$ shape memory alloy composite with temperature. Fiber and matrix of $Ti_{50}-Ni_{50}$ shape memory alloy composite are respectively $Ti_{50}-Ni_{50}$ shape memory alloy and epoxy resin. To strengthen the interfacial bonding shear stress, various surface treatments are used. They are the hand-sanded surface treatment, the acid etched surface treatment and the silane coupled surface treatment etc.. The interfacial bonding shear strength of surface treated shape memory alloy fiber is greater than that of surface untreated shape memory alloy fiber by from 10% to 16%. It is assured that the hand-sanded surface treatment and the acid etched surface treatment are the best way to strengthen the interfacial bonding shear strength of $Ti_{50}-Ni_{50}$ shape memory composite. The best treatment condition of surface is 10% HNO$_3$ solution in the etching method to strengthen the interfacial bonding shear strength of $Ti_{50}-Ni_{50}$ shape memory alloy composite.

• ETRI Journal
• /
• v.21 no.2
• /
• pp.22-30
• /
• 1999

In this paper, we derive an efficient parallel algorithm to solve the single function coarsest partition problem. This algorithm runs in O(\log2n) time using O(nlogn) operations on the EREW PRAM with O(n) memory cells used. Compared with the previous PRAM algorithms that consume O(n1+${\varepsilon}$) memory cells for some positive constant ${\varepsilon}\>0$, our algorithm consumes less memory cells without increasing the total number of operations.

• Transactions on Electrical and Electronic Materials
• /
• v.4 no.4
• /
• pp.26-29
• /
• 2003

To discover electrical properties of individual single wall nanotube(SWNT), a number of SWNT-based tubeFETs have been fabricated. The device consists of a single semiconducting SWNT on an insulating substrate, contacted at each end by metal electrodes. It presents high transconductances, and charge storage phenomenon, which is the operations of injecting electrons from the nanotube channel of a tubeFET into charge traps on the surface of the $SiO_2$ gate dielectric, thus shifting the threshold voltage. This phenomenon can be repeated many times, and maintained for the hundreds of seconds at room temperature. We will report this phenomenon as the memory effects of the SWNT, and attempt to use this property for the memory device.

• Journal of information and communication convergence engineering
• /
• v.4 no.2
• /
• pp.84-87
• /
• 2006

Lateral distributions of locally injected electrons and holes in an oxide-nitride-oxide (ONO) dielectric stack of two different silicon-oxide-nitride-oxide-silicon (SONOS) memory cells are evaluated by single-junction charge pumping technique. Spatial distribution of electrons injected by channel hot electron (CHE) for programming is limited to length of the ONO region in a locally ONO stacked cell, while is spread widely along with channel in a fully ONO stacked cell. Hot-holes generated by band-to-band tunneling for erasing are trapped into the oxide as well as the ONO stack in the locally ONO stacked cell.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.13 no.11
• /
• pp.914-920
• /
• 2000

When the charge-trap type SONOS(polysilicon-oxide-nitride-oxide-semiconductor) cells are used to flash memory, the tunneling program/erase condition to minimize the generation of interface traps was investigated. SONOSFET NVSM(Nonvolatile Semiconductor Memory) cells were fabricated using 0.35 ㎛ standard memory cell embedded logic process including the ONO cell process, based on retrograde twin-well, single-poly, single metal CMOS(Complementary Metal Oxide Semiconductor) process. The thickness of ONO triple-dielectric for the memory cell is tunnel oxide of 24 $\AA$, nitride of 74 $\AA$, blocking oxide of 25 $\AA$, respectively. The program mode(V$\_$g/=7, 8, 9 V, V$\_$s/=V$\_$d/=-3 V, V$\_$b/=floating) and the erase mode(V$\_$g/=-4, -5, -6 V, V$\_$s/=V$\_$d/=floating, V$\_$b/=3 V) by MFN(Modified Fowler-Nordheim) tunneling were used. The proposed programming condition for the flash memory of SONOSFET NVSM cells showed less degradation(ΔV$\_$th/, S, G$\_$m/) characteristics than channel MFN tunneling operation. Also, the program inhibit conditins of unselected cell for separated source lines NOR-type flash memory application were investigated. we demonstrated that the phenomenon of the program disturb did not occur at source/drain voltage of 1 V∼12 V and gate voltage of -8 V∼4 V.