wrench::ComputeService

class wrench::ComputeService : public wrench::Service

The compute service base class.

Subclassed by wrench::BareMetalComputeService, wrench::BatchComputeService, wrench::CloudComputeService, wrench::HTCondorCentralManagerService, wrench::HTCondorComputeService

Public Types

enum TerminationCause

Job termination cause enum.

Values:

enumerator TERMINATION_NONE

enumerator TERMINATION_COMPUTE_SERVICE_TERMINATED

enumerator TERMINATION_JOB_KILLED

enumerator TERMINATION_JOB_TIMEOUT

Public Functions

ComputeService(const std::string &hostname, const std::string &service_name, const std::string &scratch_space_mount_point)

Constructor.

Parameters

hostname – the name of the host on which the compute service runs
service_name – the name of the compute service
scratch_space_mount_point – the service’s scratch space’s mount point (”” if none)

std::map<std::string, double> getCoreFlopRate()

Get the per-core flop rate of the compute service’s hosts.

Throws: ExecutionException –
Returns: a list of flop rates in flop/sec

double getFreeScratchSpaceSize()

Get the free space on the compute service’s scratch storage space.

Returns: a size (in bytes)

std::vector<std::string> getHosts()

Get the list of the compute service’s compute host.

Throws

ExecutionException –
std::runtime_error –

Returns

a vector of hostnames

std::map<std::string, double> getMemoryCapacity()

Get the RAM capacities for each of the compute service’s hosts.

Throws: ExecutionException –
Returns: a map of RAM capacities, indexed by hostname

unsigned long getNumHosts()

Get the number of hosts that the compute service manages.

Throws

ExecutionException –
std::runtime_error –

Returns

the host count

std::map<std::string, double> getPerHostAvailableMemoryCapacity()

Get ram availability for each of the compute service’s host.

Throws

ExecutionException –
std::runtime_error –

Returns

the ram availability map (could be empty)

std::map<std::string, unsigned long> getPerHostNumCores()

Get core counts for each of the compute service’s host.

Throws

ExecutionException –
std::runtime_error –

Returns

a map of core counts, indexed by hostnames

std::map<std::string, unsigned long> getPerHostNumIdleCores()

Get idle core counts for each of the compute service’s host.

Throws

ExecutionException –
std::runtime_error –

Returns

the idle core counts (could be empty). Note that this doesn’t mean that asking for these cores right will mean immediate execution (since jobs may be pending and “ahead” in the queue, e.g., because they depend on current actions that are not using all available resources).

std::shared_ptr<StorageService> getScratch()

Get the compute service’s scratch storage space.

Returns: a pointer to the shared scratch space

unsigned long getTotalNumCores()

Get the total core counts for all hosts of the compute service.

Throws

ExecutionException –
std::runtime_error –

Returns

total core counts

virtual unsigned long getTotalNumIdleCores()

Get the total idle core count for all hosts of the compute service. Note that this doesn’t mean that asking for these cores right will mean immediate execution (since jobs may be pending and “ahead” in the queue, e.g., because they depend on current actions that are not using all available resources).

Throws

ExecutionException –
std::runtime_error –

Returns

total idle core count.

double getTotalScratchSpaceSize()

Get the total capacity of the compute service’s scratch storage space.

Returns: a size (in bytes)

double getTTL()

Get the time-to-live of the compute service.

Throws: ExecutionException –
Returns: the ttl in seconds

virtual bool hasScratch() const

Checks if the compute service has a scratch space.

Returns: true if the compute service has some scratch storage space, false otherwise

virtual bool isThereAtLeastOneHostWithIdleResources(unsigned long num_cores, double ram)

Method to find out if, right now, the compute service has at least one host with some idle number of cores and some available RAM. Note that this doesn’t mean that asking for these resources right will mean immediate execution (since jobs may be pending and “ahead” in the queue, e.g., because they depend on current actions that are not using all available resources).

Parameters

num_cores – the desired number of cores
ram – the desired RAM

Returns

true if idle resources are available, false otherwise

virtual void stop() override: Stop the compute service.

virtual void stop(bool send_failure_notifications, ComputeService::TerminationCause termination_cause)

Stop the compute service.

Parameters

send_failure_notifications – whether to send job failure notifications or not
termination_cause – the cause (reason) of the service’s termination

virtual void submitCompoundJob(std::shared_ptr<CompoundJob> job, const std::map<std::string, std::string> &service_specific_arguments) = 0

Method to submit a compound job to the service.

Parameters

job – The job being submitted
service_specific_arguments – the set of service-specific arguments

virtual bool supportsCompoundJobs() = 0

Returns true if the service supports pilot jobs.

Returns: true or false

virtual bool supportsPilotJobs() = 0

Returns true if the service supports compound jobs.

Returns: true or false

virtual bool supportsStandardJobs() = 0

Returns true if the service supports standard jobs.

Returns: true or false

virtual void terminateCompoundJob(std::shared_ptr<CompoundJob> job) = 0

Method to terminate a compound job.

Parameters: job – the standard job

void terminateJob(const std::shared_ptr<CompoundJob> &job)

Terminate a previously-submitted job (which may or may not be running yet)

Parameters

job – the job to terminate

Throws

std::invalid_argument –
ExecutionException –
std::runtime_error –

Public Static Attributes

static constexpr unsigned long ALL_CORES = ULONG_MAX: A convenient constant to mean “use all cores of a physical host” whenever a number of cores is needed when instantiating compute services.

static constexpr double ALL_RAM = DBL_MAX: A convenient constant to mean “use all ram of a physical host” whenever a ram capacity is needed when instantiating compute services.