Search

© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Voltage noise is the main source of dynamic variability in integrated circuits and a major concern for the design of power delivery networks (PDNs). Lower supply voltages were made possible with technology scaling, but power density was also increased. Consequently, power integrity became a key factor in the design of reliable high performance circuits. Ring oscillators clocks (ROCs) have been proposed as an alternative to mitigate the negative effects of voltage noise. However, the effectiveness highly depends on the design parameters of the PDN, power consumption patterns, and spatial locality of the ROC within the clock domain. This paper analyzes the impact of the PDN parameters and ROC location on the voltage noise and the robustness achieved by using ROCs. The capability of reacting instantaneously to large voltage droops makes ROCs an attractive solution, which also allows to relax the constraints required for the PDN design. The experiments show that up to 83% of the margins for voltage noise and up to 27% of the total leakage power can be reduced by using ROCs. In addition, PDN simplifications are possible, with fewer power interconnections or package decoupling capacitors of lower quality. Tolerance to voltage noise and related benefits can be increased with multiple ROCs., Peer Reviewed, Postprint (author's final draft)

Technology scaling enables lower supply voltages, but also increases power density of integrated circuits. In this context, power integrity becomes a major concern in the implementation of highperformance designs. This paper analyzes the influence of Ring Oscillator Clocks (ROCs) on mitigating the impacts of voltage noise. A design with an ROC as the clock source is able to work correctly even in the presence of severe and unpredictable voltage emergencies, without degrading the average performance and power metrics of the circuit. ROCs offer an instantaneous and continuous adaptation to the environment conditions, thus reducing the margins used to prevent timing failures. ROCs provide robustness independently of the power delivery network, thus relaxing the constraints required for the design of the PCB and package. As a by-product, the inherent jitter generated by ROCs produces a spreadspectrum effect that reduces electromagnetic emissions., Peer Reviewed, Postprint (published version)

Voltage noise is the main source of dynamic variability in integrated circuits and a major concern for the design of Power Delivery Networks (PDNs). Ring Oscillators Clocks (ROCs) have been proposed as an alternative to mitigate the negative effects of voltage noise as technology scales down and power density increases. However, their effectiveness highly depends on the design parameters of the PDN, power consumption patterns of the system and spatial locality of the ROCs within the clock domains. This paper analyzes the impact of the PDN parameters and ROC location on the robustness to voltage noise. The capability of reacting instantaneously to unpredictable voltage droops makes ROCs an attractive solution, which allows to reduce the amount of decoupling capacitance without downgrading performance. Tolerance to voltage noise and related benefits can be increased by using multiple ROCs and reducing the size of the clock domains. The analysis shows that up to 83% of the margins for voltage noise and up to 27% of the leakage power can be reduced by using local ROCs., Peer Reviewed, Postprint (author's final draft)

How much margin do we have to add to the delay lines of a bundled-data circuit? This paper is an attempt to give a methodical answer to this question, taking into account all sources of variability and the existing EDA machinery for timing analysis and sign-off. The paper is based on the study of the margins of a ring oscillator that substitutes a PLL as clock generator. A timing model is proposed that shows that a 12% margin for delay lines can be sufficient to cover variability in a 65nm technology. In a typical scenario, performance and energy improvements between 15% and 35% can be obtained by using a ring oscillator instead of a PLL. The paper concludes that a synchronous circuit with a ring oscillator clock shows similar benefits in performance and energy as those of bundled-data asynchronous circuits., Peer Reviewed, Postprint (author's final draft)

Manycore CMP systems are expected to grow to tens or even hundreds of cores. In this paper we show that the effective co-design of both, the network-on-chip and the coherence protocol, improves performance and power meanwhile total area resources remain bounded. We propose a snoopy-aware network-on-chip topology made of two mesh-of-tree topologies. Reducing the complexity of the coherence protocol - and hence its resources - and moving this complexity to the network, leads to a global decrease in power consumption meanwhile area is barely affected. Benefits of our proposal are due to the high-throughput and low delay of the network, but also due to the simplicity of the coherence protocol. The proposed network and protocol minimizes communication amongst cores when compared to traditional solutions based either on 2D-mesh topologies or in directory-based protocols. (C) 2015 Elsevier Ltd. All rights reserved.

Manycore CMP systems are expected to grow to tens or even hundreds of cores. In this paper we show that the effective co-design of both, the network-on-chip and the coherence protocol, improves performance and power meanwhile total area resources remain bounded. We propose a snoopy-aware network-on-chip topology made of two mesh-of-tree topologies. Reducing the complexity of the coherence protocol - and hence its resources - and moving this complexity to the network, leads to a global decrease in power consumption meanwhile area is barely affected. Benefits of our proposal are due to the high-throughput and low delay of the network, but also due to the simplicity of the coherence protocol. The proposed network and protocol minimizes communication amongst cores when compared to traditional solutions based either on 2D-mesh topologies or in directory-based protocols., Peer Reviewed, Postprint (author's final draft)

© 2015 IEEE.Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., The growing variability in nanoelectronic devices, due to uncertainties from the manufacturing process and environmental conditions (power supply, temperature, aging), requires increasing design guardbands, forcing circuits to work with conservative clock frequencies. Various schemes for clock generation based on ring oscillators and adaptive clocks have been proposed with the goal to mitigate the power and performance losses attributable to variability. However, there has been no systematic analysis to quantify the benefits of such schemes and no sign-off method has been proposed for timing correctness. This paper presents and analyzes a Reactive Clocking scheme with Variability-Tracking Jitter (RClk) that uses variability as an opportunity to reduce power by continuously adjusting the clock frequency to the varying environmental conditions, and thus, reduces guardband margins significantly. Power can be reduced between 20% and 40% at iso-performance and performance can be boosted by similar amounts at iso-power. Additionally, energy savings can be translated to substantial advantages in terms of reliability and thermal management. More importantly, the technology can be adopted with minimal modifications to conventional EDA flows., Peer Reviewed, Postprint (author's final draft)

Report - Departament Ciències de la Computació, The growing variability in nanoelectronic devices due to uncertainties from the manufacturing process and environmental conditions (power supply, temperature, aging) requires increasing design guardbands, forcing circuits to work with conservative clock frequencies. Various schemes for clock generation based on ring oscillators have been proposed with the goal to mitigate the power and performance losses attributable to variability. However, there has been no systematic analysis to quantify the benefits of such schemes.This paper presents and analyzes an Adaptive Clocking scheme with Useful Jitter (ACUJ) that uses variability as an opportunity to reduce power by adapting the clock frequency to the varying environmental conditions and, thus, reducing guardband margins significantly. Power can be reduced between 20% and 40% at iso-performance and performance can be boosted by similar amounts at iso-power. Additionally, energy savings can be translated to substantial advantages in terms of reliability and thermal management. More importantly, the technology can be adopted with minimal modifications to conventional EDA flows., Postprint (published version)

In application-specific SoCs, the irregularity of the topology ends up in a complex and customized implementation of the routing algorithm, usually relying on routing tables implemented with memory structures at source end nodes. As system size increases, the routing tables also increase in size with nonnegligible impact on power, area, and latency overheads. In this paper, we present a routing implementation for application-specific SoCs able to implement in an efficient manner (with no routing tables and using a small logic block in every switch) a deadlock-free routing algorithm in these irregular networks. The mechanism relies on a tool that maps the initial irregular topology of the SoC system into a logical regular structure where the mechanism can be applied. We provide details for both the mapping tool and the proposed routing mechanism. Evaluation results show the effectiveness of the mapping tool as well as the low area and timing requirements of the mechanism. With the mapping tool and the routing mechanism, complex irregular SoC topologies can now be supported without the need for routing tables.

© 2014 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works, The complexity of the VLSI physical design flow grows dramatically as the level of integration increases. An effective way to manage this increasing complexity is through the use of regular designs which contain more reusable parts. In this work we introduce HiReg, a new floorplanning algorithm that generates regular floorplans. HiReg automatically extracts repeating patterns in a design by using graph mining techniques. Regularity is exploited by reusing the same floorplan for multiple instances of a pattern, as long as neither area, wire length or existing hierarchy constraints are violated or compromised. The proposed scheme is targeted towards early system-level design of chip multiprocessors (CMPs). Experiments show the scalability of the method for many-core CMPs and competitive results in area and wire length, Peer Reviewed, Postprint (author's final draft)

[EN] Future chip multiprocessors will include hundreds of cores organised in a tile-based design pattern. These systems commonly employ a shared memory programming model, thus needing a coherence protocol to keep data consistent on the various levels of the cache hierarchy. Usually an invalidation-based protocol is used, where shared copies are invalidated before a write operation. In this study, the authors propose a NoC re-organisation in which a small and fast dedicated control network is used to transmit acknowledgement messages related to the invalidation process, thus relieving the NoC from a considerable percentage of traffic. The dedicated control network is evaluated both with full map directories and with a broadcast-based protocol (Hammer). Experimental evaluation shows significant gains in performance. With a low area overhead (<2.5%), the control network reduces NoC traffic and miss latency, thus reducing execution time up to 16%. Simulation results show a reduction of network traffic up to 80% and a reduction of store and load miss latency up to 70 and 40%, respectively.

[EN] It is well-known that current Chip MultiProcessor (CMP) and high-end MultiProcessor System-on-Chip (MPSoC) designs are growing in their number of components. Networks-on-Chip (NoC) provide the required connectivity for such CMP and MPSoC designs at reasonable costs. As technology advances, links become the critical component in the NoC due to their long delay and power consumption, becoming unacceptable for long global interconnects. In this paper we present a new switch architecture that reduces the negative impact of links on the NoC. We call our proposal distributed switch. The distributed switch spreads the circuitry of the switch onto the links. Thus, packets are buffered, routed, and forwarded at the same time they are crossing the link. Distributing a modular switch onto the link improves the trade off between the power consumption and the operating frequency of the entire network. On the contrary, area resources are increased. Additionally, the distributed switch presents better fault tolerance and process variation behavior with respect to a non-distributed switch.

[EN] Current integration scales allow designing chip multiprocessors (CMP), where cores are interconnected by means of a network-on-chip (NoC). Unfortunately, the small feature size of current integration scales causes some unpredictability in manufactured devices because of process variation. In NoCs, variability may affect links and routers causing them not to match the parameters established at design time. In this paper, we first analyze the way that manufacturing deviations affect the components of a NoC by applying a new comprehensive and detailed within-die variability model to 200 instances of an 8¿8 mesh NoC synthesized using 45 nm technology. Later, we show that GALS-based NoCs present communication bottlenecks under process variation which cannot be avoided by using just device-level solutions but higher level architectural approaches are required. Therefore, to overcome this performance reduction, we draft a novel architectural approach, called performance domains, intended to reduce the negative impact of variability on application execution time. This mechanism is suitable when several applications are simultaneously running in the CMP chip.

[EN] Current integration scales make possible to design chip multiprocessors with a large amount of cores interconnected by a NoC. Unfortunately, they also bring process variation, posing a new burden to processor manufacturers. Regarding the NoC, variability causes that the delays of links and routers do not match those initially established at design time. In this paper we analyze how variability affects the NoC by applying a new variability model to 100 instances of an 8 ¿ 8 mesh NoC synthesized using 45 nm technology. We also show that GALS-based NoCs present communication bottlenecks due to the slower components of the network, which cause congestion, thus reducing performance. This performance reduction finally affects the applications being executed in the CMP because they may be mapped to slower areas of the chip. In this paper we show that using a mapping algorithm that considers variability data may improve application execution time up to 50%. © 2010 Elsevier Inc. All rights reserved.

[EN] The high-performance computing domain is enriching with the inclusion of networks-on-chip (NoCs) as a key component of many-core (CMPs or MPSoCs) architectures. NoCs face the communication scalability challenge while meeting tight power, area, and latency constraints. Designers must address new challenges that were not present before. Defective components, the enhancement of application-level parallelism, or power-aware techniques may break topology regularity, thus, efficient routing becomes a challenge. This paper presents universal logic-based distributed routing (uLBDR), an efficient logic-based mechanism that adapts to any irregular topology derived from 2-D meshes, instead of using routing tables. uLBDR requires a small set of configuration bits, thus being more practical than large routing tables implemented in memories. Several implementations of uLBDR are presented highlighting the tradeoff between routing cost and coverage. The alternatives span from the previously proposed LBDR approach (with 30% of coverage) to the uLBDR mechanism achieving full coverage. This comes with a small performance cost, thus exhibiting the tradeoff between fault tolerance and performance. Power consumption, area, and delay estimates are also provided highlighting the efficiency of the mechanism. To do this, different router models (one for CMPs and one for MPSoCs) have been designed as a proof concept.

[EN] Recently, 3D stacking has been proposed to alleviate the memory bandwidth limitation arising in chip multiprocessors (CMPs). As the number of integrated cores in the chip increases the access to external memory becomes the bottleneck, thus demanding larger memory amounts inside the chip. The most accepted solution to implement vertical links between stacked dies is by using Through Silicon Vias (TSVs). However, TSVs are exposed to misalignment and random defects compromising the yield of the manufactured 3D chip. A common solution to this problem is by over-provisioning, thus impacting on area and cost. In this paper, we propose a fault-tolerant vertical link design. With its adoption, fault-tolerant vertical links can be implemented in a 3D chip design at low cost without the need of adding redundant TSVs (no over-provision). Preliminary results are very promising as the fault-tolerant vertical link design increases switch area only by 6.69% while the achieved interconnect yield tends to 100%.

[EN] As technology advances, the number of cores in Chip MultiProcessor systems and MultiProcessor Systems-on-Chips keeps increasing. The network must provide sustained throughput and ultra-low latencies. In this paper we propose new pipelined switch designs focused in reducing the switch latency. We identify the switch components that limit the switch frequency: the arbiter. Then, we simplify the arbiter logic by using multiple smaller arbiters, but increasing greatly the switch area. To solve this problem, a second design is presented where the routing traversal and arbitrations tasks are mixed. Results demonstrate a switch latency reduction ranging from 10% to 21%. Network latency is reduced in a range from 11% to 15%. © 2011 Elsevier B.V. All rights reserved.