Applying autonomic computing concepts to parallel computing using intelligent agents

The work reported in this paper is motivated by the fact that there is a need to apply autonomic computing concepts to parallel computing systems. Advancing on prior work based on intelligent cores [36], a swarm-array computing approach, this paper focuses on ‘Intelligent agents’ another swarm-array computing approach in which the task to be executed on a parallel computing core is considered as a swarm of autonomous agents. A task is carried to a computing core by carrier agents and is seamlessly transferred between cores in the event of a predicted failure, thereby achieving self-ware objectives of autonomic computing. The feasibility of the proposed swarm-array computing approach is validated on a multi-agent simulator.

Achieving intelligent agents and its feasibility in swarm-array computing

The work reported in this paper proposes ‘Intelligent Agents’, a Swarm-Array computing approach focused to apply autonomic computing concepts to parallel computing systems and build reliable systems for space applications. Swarm-array computing is a robotics a swarm robotics inspired novel computing approach considered as a path to achieve autonomy in parallel computing systems. In the intelligent agent approach, a task to be executed on parallel computing cores is considered as a swarm of autonomous agents. A task is carried to a computing core by carrier agents and can be seamlessly transferred between cores in the event of a predicted failure, thereby achieving self-* objectives of autonomic computing. The approach is validated on a multi-agent simulator.

Mechanistic insights into a gas–solid reaction in molecular crystals: the role of hydrogen bonding

Reactions in (molecular) organic crystalline solids have been shown to be important for exerting control that is unattainable over chemical transformations in solution. Such control has also been achieved for reactions within metal– organic cages. In these examples, the reactants are already in place within the crystals following the original crystal growth. The post-synthetic modification of metal–organic frameworks (MOFs and indeed reactions and catalysis within MOFs have been recently demonstrated; in these cases the reactants enter the crystals through permanent channels. Another growing area of interest within molecular solid-state chemistry is synthesis by mechanical co-grinding of solid reactants—often referred to as mechanochemistry. Finally, in a small number of reported examples, molecules also have been shown to enter nonporous crystals directly from the gas or vapor phase, but in only a few of these examples does a change in covalent bonding result, which indicates that a reaction occurs within the nonporous crystals. It is this latter type of highly uncommon reaction that is the focus of the present study.

Self-assembled healable materials through a combination of pi-pi stacking and hydrogen bonding

The discovery of polymers with stimuli responsive physical properties is a rapidly expanding area of research. At the forefront of the field are self-healing polymers, which, when fractured can regain the mechanical properties of the material either autonomically, or in response to a stimulus. It has long been known that it is possible to promote healing in conventional thermoplastics by heating the fracture zone above the Tg of the polymer under pressure. This process requires reptation and subsequent re-entanglement of macromolecules across the fracture void, which serves to bridge, and ‘heal’ the crack. The timescale for this mechanism is highly dependent on the molecular weight of the polymer being studied. This process is in contrast to that required to affect healing in supramolecular polymers such as the plasticised, hydrogen bonded elastomer reported by Leibler et al. The disparity in bond energies between the non-covalent and covalent bonds within supramolecular polymers results in fractures propagating through scission of the comparatively weak supramolecular interactions, rather than through breaking the stronger, covalent bonds. Thus, during the healing process the macromolecules surrounding the fracture site only need sufficient energy to re-engage their supramolecular interactions in order to regenerate the strength of the pristine material. Herein we describe the design, synthesis and optimization of a new class of supramolecular polymer blends that harness the reversible nature of pi-pi stacking and hydrogen bonding interactions to produce self-supporting films with facile healable characteristics.

Hydrogen bonded supramolecular elastomers : correlating hydrogen bonding strength with morphology and rheology

A series of six low molecular weight elastomers with hydrogen bonding end-groups have been designed, synthesised and studied. The poly(urethane) based elastomers all contained essentially the same hard block content (ca. 11%) and differ only in the nature of their end-groups. Solution state 1H NMR spectroscopic analysis of model compounds featuring the end-groups demonstrate that they all exhibit very low binding constants, in the range 1.4 to 45.0 M-1 in CDCl3, yet the corresponding elastomers each possess a markedly different nanoscale morphology and rheology in the bulk. We are able to correlate small variations of the binding constant of the end-groups with dramatic changes in the bulk properties of the elastomers. These results provide an important insight into the way in which weak non-covalent interactions can be utilized to afford a range of self-assembled polyurethane based materials that feature different morphologies.

Electronic transitions and bonding properties in a series of five-coordinate “16-electron” complexes [Mn(CO)3(L2)]− (L2=chelating redox-active π-donor ligand)

In this article we present for the first time accurate density functional theory (DFT) and time-dependent (TD) DFT data for a series of electronically unsaturated five-coordinate complexes [Mn(CO)(3)(L-2)](-), where L-2 stands for a chelating strong pi-donor ligand represented by catecholate, dithiolate, amidothiolate, reduced alpha-diimine (1,4-dialkyl-1,4-diazabutadiene (R-DAB), 2,2'-bipyridine) and reduced 2,2'-biphosphinine types. The single-crystal X-ray structure of the unusual compound [Na(BPY)][Mn(CO)(3)(BPY)]center dot Et2O and the electronic absorption spectrum of the anion [Mn(CO)(3)(BPY)](-) are new in the literature. The nature of the bidentate ligand determines the bonding in the complexes, which varies between two limiting forms: from completely pi-delocalized diamagnetic {(CO)(3)Mn-L-2}(-) for L-2 = alpha-diimine or biphosphinine, to largely valence-trapped {(CO)(3)Mn-1-L-2(2-)}(-) for L-2(2-) = catecholate, where the formal oxidation states of Mn and L-2 can be assigned. The variable degree of the pi-delocalization in the Mn(L-2) chelate ring is indicated by experimental resonance Raman spectra of [Mn(CO)(3)(L-2)](-) (L-2=3,5-di-tBu-catecholate and iPr-DAB), where accurate assignments of the diagnostically important Raman bands have been aided by vibrational analysis. The L-2 = catecholate type of complexes is known to react with Lewis bases (CO substitution, formation of six-coordinate adducts) while the strongly pi-delocalized complexes are inert. The five-coordinate complexes adopt usually a distorted square pyramidal geometry in the solid state, even though transitions to a trigonal bipyramid are also not rare. The experimental structural data and the corresponding DFT-computed values of bond lengths and angles are in a very good agreement. TD-DFT calculations of electronic absorption spectra of the studied Mn complexes and the strongly pi-delocalized reference compound [Fe(CO)(3)(Me-DAB)] have reproduced qualitatively well the experimental spectra. Analyses of the computed electronic transitions in the visible spectroscopic region show that the lowest-energy absorption band always contains a dominant (in some cases almost exclusive) contribution from a pi(HOMO) -> pi*(LUMO) transition within the MnL2 metallacycle. The character of this optical excitation depends strongly on the composition of the frontier orbitals, varying from a partial L-2 -> Mn charge transfer (LMCT) through a fully delocalized pi(MnL2) -> pi*(MnL2) situation to a mixed (CO)Mn -> L-2 charge transfer (LLCT/MLCT). The latter character is most apparent in the case of the reference complex [Fe(CO)(3)(Me-DAB)]. The higher-lying, usually strongly mixed electronic transitions in the visible absorption region originate in the three lower-lying occupied orbitals, HOMO - 1 to HOMO - 3, with significant metal-d contributions. Assignment of these optical excitations to electronic transitions of a specific type is difficult. A partial LLCT/MLCT character is encountered most frequently. The electronic absorption spectra become more complex when the chelating ligand L-2, such as 2,2'-bipyridine, features two or more closely spaced low-lying empty pi* orbitals.

A new interpretation of the bonding properties and UV–vis spectra of [M3(CO)12] clusters (M = Ru, Os): a TD-DFT study

DFT and TD-DFT calculations (ADF program) were performed in order to analyze the electronic structure of the [M-3(CO)(12)] clusters (M = Ru, Os) and interpret their electronic spectra. The highest occupied molecular orbitals are M-M bonding (sigma) involving different M-M bonds, both for Ru and Os. They participate in low-energy excitation processes and their depopulation should weaken M-M bonds in general. While the LUMO is M-NI and M-CO anti-bonding (sigma*), the next, higher-lying empty orbitals have a main contribution from CO (pi*) and either a small (Ru) or an almost negligible one (Os) from the metal atoms. The main difference between the two clusters comes from the different nature of these low-energy unoccupied orbitals that have a larger metal contribution in the case of ruthenium. The photochemical reactivity of the two clusters is reexamined and compared to earlier interpretations.

Handling single node failures using agents in computer clusters

The work reported in this paper is motivated towards handling single node failures for parallel summation algorithms in computer clusters. An agent based approach is proposed in which a task to be executed is decomposed to sub-tasks and mapped onto agents that traverse computing nodes. The agents intercommunicate across computing nodes to share information during the event of a predicted node failure. Two single node failure scenarios are considered. The Message Passing Interface is employed for implementing the proposed approach. Quantitative results obtained from experiments reveal that the agent based approach can handle failures more efficiently than traditional failure handling approaches.

Exploring carrier agents in swarm-array computing

How can a bridge be built between autonomic computing approaches and parallel computing systems? The work reported in this paper is motivated towards bridging this gap by proposing a swarm-array computing approach based on ‘Intelligent Agents’ to achieve autonomy for distributed parallel computing systems. In the proposed approach, a task to be executed on parallel computing cores is carried onto a computing core by carrier agents that can seamlessly transfer between processing cores in the event of a predicted failure. The cognitive capabilities of the carrier agents on a parallel processing core serves in achieving the self-ware objectives of autonomic computing, hence applying autonomic computing concepts for the benefit of parallel computing systems. The feasibility of the proposed approach is validated by simulation studies using a multi-agent simulator on an FPGA (Field-Programmable Gate Array) and experimental studies using MPI (Message Passing Interface) on a computer cluster. Preliminary results confirm that applying autonomic computing principles to parallel computing systems is beneficial.

An MPI-based implementation of intelligent agents on clusters

Recent research in multi-agent systems incorporate fault tolerance concepts, but does not explore the extension and implementation of such ideas for large scale parallel computing systems. The work reported in this paper investigates a swarm array computing approach, namely 'Intelligent Agents'. A task to be executed on a parallel computing system is decomposed to sub-tasks and mapped onto agents that traverse an abstracted hardware layer. The agents intercommunicate across processors to share information during the event of a predicted core/processor failure and for successfully completing the task. The feasibility of the approach is validated by implementation of a parallel reduction algorithm on a computer cluster using the Message Passing Interface.

Intelligent agents for fault tolerance: from multi-agent simulation to cluster-based implementation

Recent research in multi-agent systems incorporate fault tolerance concepts, but does not explore the extension and implementation of such ideas for large scale parallel computing systems. The work reported in this paper investigates a swarm array computing approach, namely 'Intelligent Agents'. A task to be executed on a parallel computing system is decomposed to sub-tasks and mapped onto agents that traverse an abstracted hardware layer. The agents intercommunicate across processors to share information during the event of a predicted core/processor failure and for successfully completing the task. The feasibility of the approach is validated by simulations on an FPGA using a multi-agent simulator, and implementation of a parallel reduction algorithm on a computer cluster using the Message Passing Interface.

Principal-agent issues in asset acquisition: UK institutions and their investment agents

This paper explores principal‐agent issues in the stock selection processes of institutional property investors. Drawing upon an interview survey of fund managers and acquisition professionals, it focuses on the relationships between principals and external agents as they engage in property transactions. The research investigated the extent to which the presence of outcome‐based remuneration structures could lead to biased advice, overbidding and/or poor asset selection. It is concluded that institutional property buyers are aware of incentives for opportunistic behaviour by external agents, often have sufficient expertise to robustly evaluate agents’ advice and that these incentives are counter‐balanced by a number of important controls on potential opportunistic behaviour. There are strong counter‐incentives in the need for the agents to establish personal relationships and trust between themselves and institutional buyers, to generate repeat and related business and to preserve or generate a good reputation in the market.

Action observation and robotic agents: learning and anthropomorphism

The ‘action observation network’ (AON), which is thought to translate observed actions into motor codes required for their execution, is biologically tuned: it responds more to observation of human, than non-human, movement. This biological specificity has been taken to support the hypothesis that the AON underlies various social functions, such as theory of mind and action understanding, and that, when it is active during observation of non-human agents like humanoid robots, it is a sign of ascription of human mental states to these agents. This review will outline evidence for biological tuning in the AON, examining the features which generate it, and concluding that there is evidence for tuning to both the form and kinematic profile of observed movements, and little evidence for tuning to belief about stimulus identity. It will propose that a likely reason for biological tuning is that human actions, relative to non-biological movements, have been observed more frequently while executing corresponding actions. If the associative hypothesis of the AON is correct, and the network indeed supports social functioning, sensorimotor experience with non-human agents may help us to predict, and therefore interpret, their movements.