Search

This paper investigated DRAM DIMM errors using field records in replacement network servers. Large DRAM samples of about 40 K were collected over a 2.5 years period from 23 different server types, included various DIMMs from three different DRAM manufacturers with densities between 4 and 128 GB, and speeds between 1066 and 2400 Mbps. Errors that occurred during system operation were classified as either correctable (CE) or uncorrectable (UE) errors based on error correction code (ECC) schemes built into the servers. Of the collected DIMMS, 24% had recorded errors, where CE-only, UE-only, and UE and CE together comprised 28%, 43%, and 29% of recorded errors, respectively. Since UEs can cause large-scale failures, systems are replaced upon any UE occurrence. Approximately half UE-only DIMMs had 1 UE error. In contrast, many DIMMs had billions of CE errors, where a faulty location may be repetitively accessed. Such drastic differences in UE and CE counts help explain the importance of ECC and error mitigation schemes. Comparative analyses of errors were made over the manufacturers and operating speeds. After reasonable adjustments for repetitive counts of errors, failure in time (FIT) differences were up to 38% over manufacturers. Higher speed DIMMs generally had higher FIT with 2400 Mbps DIMMs exhibiting 6.7 times FIT of 1066 Mbps DIMMs. [ABSTRACT FROM AUTHOR]

This paper analyzes signal distortion caused by nanometer-scale solder ball fractures. A solder ball fracture causes an intermittent open circuit on the transmission line. The resulting basic failure mechanism is a drop in signal voltage, due to the capacitance-induced Alternating Current (AC)-coupling effect (which is induced by the fractured solder ball). The two major contributing factors to this error are fracture height and the persisting duration of the consecutive same-logic-value signal. The required signal that induces a voltage drop, sufficient to detect nanometer-scale solder ball fractures, can be composed by repetition of certain signal patterns even for the I/O connections with run-length restrictions. The methodology is newly proposed to determine potential ranges of solder ball fractures. Test pattern generation is made by maximally exploiting the compounding effect of various sizes of same data bits to generate effective run-length that is larger than maximum run length for the purpose of detecting intermittent solder ball fractures. In DDR3 memory systems with 5-nm solder ball fractures, at least 29 bits of consecutive logic “1” or “0” signals are required to detect fractures. If the system has a maximum run-length of 10, 20, or 30 bits, the test signal—which has the equivalent voltage-dropping effect as 29-consecutive bits—can be generated with six, two, or one repetition of the test pattern, respectively; the test pattern is in the form of concatenated N-1 bits of consecutive logic “1” and 1 bit of logic “0” where N is the maximum run length. In addition, a SPICE simulation was conducted to confirm correlation between calculations and actual results. In the simulation, in order to detect a 5-nm solder ball fracture, at least 37 bits of consecutive signal were required. [ABSTRACT FROM AUTHOR]

This paper analyzes the solder ball fracture that could be a source of intermittent errors. The electrical characteristics of a momentary fracture (open), which may appear at the very beginning of a progressive solder ball failure, are extensively studied. The alternating current (AC)-coupling capacitor is defectively formed because of the fractured solder ball, and it can block low-frequency components of the bit data stream. The distorted signal reduces signal integrity under the fracture and results in increased jitter and reduced eye window. This fracture causes the dropping of the signal voltage level, and this dropping erroneously affects the system when multiple failing conditions are simultaneously satisfied. The major failing condition is the fracture height size, which determines the defective AC-coupling capacitor and changes the channel transmission characteristic. The other major condition is the bit pattern, which includes the frequencies affected by the defective channel. SPICE simulation is conducted to demonstrate the effects of a momentary fracture using the DDR3 memory tester system. In the case of a 10 nm fractured solder ball with a pseudo-random binary sequence (PRBS) pattern, the eye height is reduced from 597 mV to 349 mV, and the jitter is increased from 38 ps to 132 ps. The bits that violate the eye-mask window begin to appear with a heavy bit stream and cause intermittent bit errors. [ABSTRACT FROM AUTHOR]

This paper proposes a new method of investigating the effect of void or fracture in FBGA solder ball on the DDR4 data signal rise time and inter-symbol interference (ISI), by loading the data line with a capacitor. A void or fracture in solder ball increases its capacitance which effects the data signal rise time and increases ISI. For measuring ISI large consecutive patterns of 1's or 0's followed by a changing bit are used. However in in-field systems it is not possible to run large patterns of 1's or 0's. So the data line is loaded with a 0.2 pF capacitive load on a UDIMM test card to mimic the increased capacitance due to FBGA solder ball void defect of height 0.2 mm and cross sectional area of 0.0045 mm2. The loaded line shows increase in rise time of 16 ps. For loaded line the data eye opening is 0.077 UI lesser. This decrease in data eye means more ISI and will cause increase in intermittent errors. • Proposes a new method of investigating the effect of void or fracture in FBGA solder ball • A capacitor is loaded on the DDR4 data line to mimic the effect of FBGA fault • Rise time of the data signal with capacitive load shows increase • Channel loss of the data line with capacitive load shows increase hence more intermittent errors [ABSTRACT FROM AUTHOR]

We present here energy straggling data for alpha particles from 241Am in air and its constituents, and an experimental investigation of Bragg's rule. High resolution alpha spectroscopy system from ORTEC with an ultra-thin source, an ion implanted thin entrance window detector of small diameter, and state of the art counting electronics with a 4096 MCA is used. In the controlled experimental environment gases of 5N purity are used to measure energy straggling. Measured values are compared with the theoretical predictions of Bohr, Bethe-Livingston, Titeica and Yang et. al. An experimental investigation on using Bragg's rule to estimate the energy straggling in air from the experimental obtained data of constituents is presented. • High-resolution spectroscopy for energy straggling of 5.486 MeV alpha particles in air and constituents. • A promising experimental verification of Bragg's rule. • Energy straggling data in air is estimated from those of measured in its constituents. • An extensive comparison of experimental data against theoretical predictions. [ABSTRACT FROM AUTHOR]

As process technology shrinks down, the distances among storage cells in DRAMs gets smaller. Smaller distances among cells cause the cell-to-cell interference to increase. Due to the proximity to neighboring cells, a DRAM cell loses the stored charge in the cell faster with repetitive accesses on an adjacent row. Such phenomenon is known as row hammering failure and has been reported to occur in the commodity DDR3 components manufactured with the process technologies under 3 × nm technology. This work developed a statistical model of row hammering failures based on experimental results obtained with commodity DDR3 SDRAMs of 3 × nm technology. The statistical distribution for the failing rows with respect to the number of hammerings matched the normal distribution. The means μ HMR and standard deviations σ HMR of the number of hammerings that cause row hammering failure were apparently different among three different manufacturers. The means of the manufacturers varied by more than 200% and could be sufficiently used to characterize the reliability of the device from a row hammering stress perspective. Based on the derived statistical model, the failed parts-per-million (ppm) was calculated to give, on average, 164.6, 82.6 and 22.2, respectively, for the manufacturers. [ABSTRACT FROM AUTHOR]

This paper shares the effects of row hammer fault through high-energy proton radiation. The significance of row hammer fault is highlighted in terms of two different technologies in DDR4 SDRAM. Row hammer stress prevents nearby storage cells from maintaining storage data within the retention time of 64-ms. The stress is worsened by the radiation damage in silicon due to the high-energy particles. Proton-based radiation damage tests were performed with DDR4 SDRAM components from two different technologies. Experiment results showed that after proton irradiation, the number of bit errors caused by the row hammering test increased about 41% and 66% in technologies 2x-nm and 2y-nm, respectively. With 2y-nm technology, bit errors started to appear from 5 K Number of Hammering (N HMR )—the equivalent of 500-μs retention time. For 2y-nm components, more than 90% of failed words had multiple failed cells, which could not be corrected by Single Error Correction and Double Error Detection (SEC-DED) codes. [ABSTRACT FROM AUTHOR]

To evaluate a device sensitivity against alpha particles, traditional Single Event Effect (SEE) tests are conducted using isotope source, which emits particles just above 5 MeV. Relentless downscaling and higher packing density have driven the demand of developing increasingly complex packaging: smaller, thinner, having enormous input/output pins count per chip. Flip-chip bonded devices meet all these demands, but their testing against alpha particles is a big challenge. The range of alpha ions, emitted by the isotope sources, is very short, which precludes their penetration till active circuit – from either side of the chip. This paper presents an evidence that high energy alpha irradiation can potentially be used to measure and correlate alpha SEE cross-section for such devices. The proposed method uses high energy alpha particles, directed from the backside of die, to mimic low energy (~ 5 MeV) at the sensitive volume (SV). The incident particles penetrate the entire silicon substrate and deposit charge in the SV to induce upsets. The energy and LET of an ion at the SV, having traversed the entire substrate, is determined using TRIM. SEE experiments are performed on 14 nm FinFET SRAM devices, assembled in flip-chip and wire-bonded structures, respectively, for backside and traditional top-side testing. High energy alpha irradiations were simulated using CRÈME-MC – a Geant4 based Monte Carlo transport code. Tests and simulation results, for traditional and proposed methods, are presented for correlation. [ABSTRACT FROM AUTHOR]

This paper investigates the failure mechanism manifested in DDR3 SDRAMs under 3 × nm technology. DRAM cells should retain the stored value if they are refreshed within the cell retention time of 64 ms at minimum. However the charge in a DRAM cell leaked faster, and the values of the stressed cells could not be retained with valid yet stressful hammered accesses to a row. An experiment of accelerated discharging by hammered accesses was duplicated by a SPICE simulation with a TCAD device model of a DRAM cell. Experiments with commercial DDR3 discrete components from three major memory manufacturers were performed to confirm the validity of the SPICE simulation. The contributions of each in triggering and accelerating the failure mechanisms are investigated depending on the three test parameters—t RP , data pattern, and temperature—based on the experimental results. In the experiments, all commercial DDR3 components failed much earlier than the specified limit of allowed accesses. In the worst condition, the failure in a normal cell of a component occurred at 200 K, which is 15.23% of the permitted cell retention time. [ABSTRACT FROM AUTHOR]

A theoretical and experimental investigation of Bragg's rule in low energy loss region where charge exchange straggling can be ignored is presented. Energy-loss straggling of 5.486 MeV alpha particles in air and 5N purity constituent gases was measured using high-resolution alpha spectroscopy system. Measured values are compared with Bohr, Bethe-Livingston and Yang et al. formulations. An experimental investigation of Bragg's rule to estimate the energy-loss straggling in air using the experimental data of constituents is discussed. • High-resolution spectroscopy for energy straggling alpha particles in air and constituents for low mean energy loss. • Experimental and theoretical energy loss straggling investigations adopting Bragg's rule. • An experimental verification of Bragg's rule to estimate energy straggling in air from the data of constituents. [ABSTRACT FROM AUTHOR]