Utilize multiple cores for Pandas

I have the following Python (2.7) function using Pandas which I need to run on 400 GB. A test run on 150 GB took 4 hours to complete successfully (memory on machine is 128 GB and 16 cores, 4TB disk). Is there a smart way to utilize all the cores on my CPU for this and doing this action in parallel to decrease the process time:

def create_data(headers, filters, filesFolder, remove_chars, outputFolder):
for ds in headers:
    for data_file in glob.glob(filesFolder + '*' + ds + '*.csv'):
        x=0
        for data in pd.read_csv(data_file, sep = '€', names = headers[ds], engine = 'python', chunksize = 10000000):
            logger.info('Pandas Read %s' %(str(x) + '_' + os.path.basename(data_file)) )
            for fil in filters[ds]:
                try:
                    data[fil] = data[fil].astype('O')
                    data = data[data[fil] == filters[ds][fil]]
                    logger.info('Filtered data for %s succesfully. Filters are %s' %(str(x) + '_' + os.path.basename(data_file) , filters[ds]))
                except:
                    logger.info('Could not apply filter %s for %s' %(filters[ds][fil], str(x) + '_' + os.path.basename(data_file) ))
                    pass
            for char in remove_chars:   
                try:
                    data = data.replace({char: ''}, regex=True)
                    logger.info('Removed char %s from %s succesfully' %(char, str(x) + '_' + os.path.basename(data_file) ))
                except:
                    logger.info('Could not remove char %s from %s' %(char, str(x) + '_' + os.path.basename(data_file) ))
                    pass
            try:
                data.to_csv(outputFolder + ds + '/' + str(x) + '_' + os.path.basename(data_file).split('.')[0].strip() + '.csv', enconding = 'utf-8')
                logger.info('Saved csv %s succesfully' %(str(x) + '_' + os.path.basename(data_file)))
                x+=1
            except:
                logger.info('Failed to Save %s' %(str(x) + '_' + os.path.basename(data_file)))