Andreas Hansson

A unified approach to constrained mapping and routing on network-on-chip architectures

One of the key steps in Network-on-Chip (NoC) based design is spatial mapping of cores and routing of the communication between those cores. Known solutions to the mapping and routing problem first map cores onto a topology and then route communication, using separated and possibly conflicting objective functions. In this paper we present a unified single-objective algorithm, called Unified MApping, Routing and Slot allocation (UMARS). As the main contribution we show how to couple path selection, mapping of cores and TDMA time-slot allocation such that the network required to meet the constraints of the application is minimized. The time-complexity of UMARS is low and experimental results indicate a run-time only 20% higher than that of path selection alone. We apply the algorithm to an MPEG decoder System-on-Chip (SoC), reducing area by 33%, power by 35% and worst-case latency by a factor four over a traditional multi-step approach.

A Unified Approach to Mapping and Routing on a Network-on-Chip for Both Best-Effort and Guaranteed Service Traffic

One of the key steps in Network-on-Chip-based design is spatial mapping of cores and routing of the communication between those cores. Known solutions to the mapping and routing problems first map cores onto a topology and then route communication, using separate and possibly conflicting objective functions. In this paper, we present a unified single-objective algorithm, called Unified MApping, Routing, and Slot allocation (UMARS+). As the main contribution, we show how to couple path selection, mapping of cores, and channel time-slot allocation to minimize the network required to meet the constraints of the application. The time-complexity of UMARS+ is low and experimental results indicate a run-time only 20% higher than that of path selection alone. We apply the algorithm to an MPEG decoder System-on-Chip, reducing area by 33%, power dissipation by 35%, and worst-case latency by a factor four over a traditional waterfall approach.

NADPH-dependent thioredoxin reductase and 2-Cys peroxiredoxins are needed for the protection of Mg–protoporphyrin monomethyl ester cyclase

The chloroplast‐localized NADPH‐dependent thioredoxin reductase (NTRC) has been found to be able to reduce hydrogen peroxide scavenging 2‐Cys peroxiredoxins. We show that the Arabidopsis ntrc mutant is perturbed in chlorophyll biosynthesis and accumulate intermediates preceding protochlorophyllide formation. A specific involvement of NTRC during biosynthesis of protochlorophyllide is indicated from in vitro aerobic cyclase assays in which the conversion of Mg–protoporhyrin monomethyl ester into protochlorophyllide is stimulated by addition of the NTRC/2‐Cys peroxiredoxin system. These findings support the hypothesis that this NADPH‐dependent hydrogen peroxide scavenging system is particularly important during periods with limited reducing power from photosynthesis, e.g. under chloroplast biogenesis.

Three semidominant barley mutants with single amino acid substitutions in the smallest magnesium chelatase subunit form defective AAA+ hexamers

Many enzymes of the bacteriochlorophyll and chlorophyll biosynthesis pathways have been conserved throughout evolution, but the molecular mechanisms of the key steps remain unclear. The magnesium chelatase reaction is one of these steps, and it requires the proteins BchI, BchD, and BchH to catalyze the insertion of Mg2+ into protoporphyrin IX upon ATP hydrolysis. Structural analyses have shown that BchI forms hexamers and belongs to the ATPases associated with various cellular activities (AAA+) family of proteins. AAA+ proteins are Mg2+-dependent ATPases that normally form oligomeric ring structures in the presence of ATP. By using ATPase-deficient BchI subunits, we demonstrate that binding of ATP is sufficient to form BchI oligomers. Further, ATPase-deficient BchI proteins can form mixed oligomers with WT BchI. The formation of BchI oligomers is not sufficient for magnesium chelatase activity when combined with BchD and BchH. Combining WT BchI with ATPase-deficient BchI in an assay disrupts the chelatase reaction, but the presence of deficient BchI does not inhibit ATPase activity of the WT BchI. Thus, the ATPase of every WT segment of the hexamer is autonomous, but all segments of the hexamer must be capable of ATP hydrolysis for magnesium chelatase activity. We suggest that ATP hydrolysis of each BchI within the hexamer causes a conformational change of the hexamer as a whole. However, hexamers containing ATPase-deficient BchI are unable to perform this ATP-dependent conformational change, and the magnesium chelatase reaction is stalled in an early stage.

Composability and Predictability for Independent Application Development,Verification, and Execution

System-on-chip (soc) design gets increasingly complex, as a growing number of applications are integrated in modern systems. Some of these applications have real-time requirements, such as a minimum throughput or a maximum latency. To reduce cost, system resources are shared between applications, making their timing behavior inter-dependent. Real-time requirements must hence e verified for all possible combinations of concurrently executing applications, which is not feasible with commonly used simulation-based techniques. This chapter addresses this problem using two complexity-reducing concepts: composability and predictability. Applications in a composable system are completely isolated and cannot affect each others behaviors, enabling them to be independently verified. Predictable systems, on the other hand, provide lower bounds on performance, allowing applications to be verified using formal performance analysis. Five techniques to achieve composability and/or predictability in soc resources are presented and we explain their implementation for processors, interconnect, and memories in our platform.

Design and implementation of an operating system for composable processor sharing

Multi-Processor Systems on Chip (MPSoC) run multiple independent applications, often developed by different parties. The applications share the hardware resources, e.g. processors, memories and interconnect. The sharing typically causes interference between the applications, which severely complicates system integration and verification. Even if the applications are verified in isolation, the system designer must verify the combined behaviour, leading to an explosion in design complexity. Composable MPSoCs have no interference between applications, thus allowing independent design and verification. For an MPSoC to be composable, all the hardware resources must offer composability. A particularly challenging resource is the processors, often purchased as off-the-shelf intellectual property. In this work we present the design and implementation of CompOSe, a light-weight (only 1500 lines of code) composable operating system for MPSoCs. CompOSe uses fixed-size time slices, coupled with a composable scheduler, to enable composable processor sharing. Using instances of ARM7, ARM11 and the Xilinx MicroBlaze we experimentally demonstrate the ability to provide temporal composability, even in the presence of dynamic application behaviour and multiple use cases. We do so using a diverse set of processor architectures, without requiring any hardware modifications. We also show how CompOSe allows slack to be distributed within and between applications through a novel two-level scheduler and slack-distribution system.

Channel trees: reducing latency by sharing time slots in time-multiplexed networks on chip

Networks on Chip (NoC) have emerged as the design paradigm for scalable System on Chip communication infrastructure. A growing number of applications, often with firm (FRT) or soft real-time (SRT) requirements, are integrated on the same chip. To provide time-related guarantees, NoC resources are reserved, e.g. by non-work-conserving time-division multiplexing (TDM). Traditionally, reservations are made on a per-communication-channel basis, thus providing FRT guarantees to individual channels. For SRT applications, this strategy is overly restrictive, as slack bandwidth is not used to improve performance. In this paper we introduce the concept of channel trees, where time slots are reserved for sets of communication channels. By employing work-conserving arbitration within a tree, we exploit the inherent single-threaded behaviour of the resource at the root of the tree, resulting in a drastic reduction in both average-case latency and TDM-table size. We show how channel trees enable us to halve the latter in a car entertainment SoC, and reduce the average latency by as much as much as 52% in a mobile phone SoC. By applying channel trees to an H264 decoder SoC, we increase processor utilisation by 25%.

The activity of barley NADPH-dependent thioredoxin reductase C is independent of the oligomeric state of the protein: tetrameric structure determined by cryo-electron microscopy

Thioredoxin and thioredoxin reductase can regulate cell metabolism through redox regulation of disulfide bridges or through removal of H(2)O(2). These two enzymatic functions are combined in NADPH-dependent thioredoxin reductase C (NTRC), which contains an N-terminal thioredoxin reductase domain fused with a C-terminal thioredoxin domain. Rice NTRC exists in different oligomeric states, depending on the absence or presence of its NADPH cofactor. It has been suggested that the different oligomeric states may have diverse activity. Thus, the redox status of the chloroplast could influence the oligomeric state of NTRC and thereby its activity. We have characterized the oligomeric states of NTRC from barley (Hordeum vulgare L.). This also includes a structural model of the tetrameric NTRC derived from cryo-electron microscopy and single-particle reconstruction. We conclude that the tetrameric NTRC is a dimeric arrangement of two NTRC homodimers. Unlike that of rice NTRC, the quaternary structure of barley NTRC complexes is unaffected by addition of NADPH. The activity of NTRC was tested with two different enzyme assays. The N-terminal part of NTRC was tested in a thioredoxin reductase assay. A peroxide sensitive Mg-protoporphyrin IX monomethyl ester (MPE) cyclase enzyme system of the chlorophyll biosynthetic pathway was used to test the catalytic ability of both the N- and C-terminal parts of NTRC. The different oligomeric assembly states do not exhibit significantly different activities. Thus, it appears that the activities are independent of the oligomeric state of barley NTRC.

Knock-out of the chloroplast-encoded PSI-J subunit of photosystem I in Nicotiana tabacum.

The plastid‐encoded psaJ gene encodes a hydrophobic low‐molecular‐mass subunit of photosystem I (PSI) containing one transmembrane helix. Homoplastomic transformants with an inactivated psaJ gene were devoid of PSI‐J protein. The mutant plants were slightly smaller and paler than wild‐type because of a 13% reduction in chlorophyll content per leaf area caused by an ≈ 20% reduction in PSI. The amount of the peripheral antenna proteins, Lhca2 and Lhca3, was decreased to the same level as the core subunits, but Lhca1 and Lhca4 were present in relative excess. The functional size of the PSI antenna was not affected, suggesting that PSI‐J is not involved in binding of light‐harvesting complex I. The specific PSI activity, measured as NADP+ photoreduction in vitro, revealed a 55% reduction in electron transport through PSI in the mutant. No significant difference in the second‐order rate constant for electron transfer from reduced plastocyanin to oxidized P700 was observed in the absence of PSI‐J. Instead, a large fraction of PSI was found to be inactive. Immunoblotting analysis revealed a secondary loss of the luminal PSI‐N subunit in PSI particles devoid of PSI‐J. Presumably PSI‐J affects the conformation of PSI‐F, which in turn affects the binding of PSI‐N. This together renders a fraction of the PSI particles inactive. Thus, PSI‐J is an important subunit that, together with PSI‐F and PSI‐N, is required for formation of the plastocyanin‐binding domain of PSI. PSI‐J is furthermore important for stability or assembly of the PSI complex.

Spontaneous Low-Frequency Oscillations in Cerebral Vessels: Applications in Carotid Artery Disease and Ischemic Stroke

The etiology behind and physiological significance of spontaneous oscillations in the low-frequency spectrum in both systemic and cerebral vessels remain unknown. Experimental studies have proposed that spontaneous oscillations in cerebral blood flow reflect impaired cerebral autoregulation (CA). Analysis of CA by measurement of spontaneous oscillations in the low-frequency spectrum in cerebral vessels might be a useful tool for assessing risk and investigating different treatment strategies in carotid artery disease (CAD) and stroke. We reviewed studies exploring spontaneous oscillations in the low-frequency spectrum in patients with CAD and ischemic stroke, conditions known to involve impaired CA. Several studies have reported changes in oscillations after CAD and stroke after surgery and over time compared with healthy controls. Phase shift in the frequency domain and correlation coefficients in the time domain are the most frequently used parameters for analyzing spontaneous oscillations in systemic and cerebral vessels. At present, there is no gold standard for analyzing spontaneous oscillations in the low-frequency spectrum, and simplistic models of CA have failed to predict or explain the spontaneous oscillation changes found in CAD and stroke studies. Near-infrared spectroscopy is suggested as a future complementary tool for assessing changes affecting the cortical arterial system.